Comment from the Federal Trade Commission on Artificial Intelligence and Copyright

UNITED STATES COPYRIGHT OFFICE
______
Artificial Intelligence and Copyright Docket No. 2023-6
COMMENT OF
THE UNITED STATES FEDERAL TRADE COMMISSION
October 30, 2023

FTC Comment
October 30, 2023
2
I. Introduction
The United States Federal Trade Commission (FTC) commends the Copyright Office’s
ongoing efforts to examine the copyright law and policy issues raised by artificial intelligence
(AI) technology, including the scope of copyright in works generated using AI tools and the use
of copyrighted materials in AI training. The FTC submits this comment to: (1) explain its interest
and expertise in promoting fair competition and protecting Americans from unfair or deceptive
practices in an economy in which AI is being rapidly deployed; (2) identify several issues raised
by generative AI that implicate competition and consumer protection policy, as well as copyright
policy; and (3) augment the record in this proceeding with a brief summary and full transcript of
an October 4, 2023 FTC roundtable at which creative professionals described the effects of
generative AI on their work.
II. The FTC’s Interest in AI
The FTC is an independent federal agency charged with promoting competition and
protecting consumers. Our authority derives primarily from the FTC Act, which prohibits unfair
or deceptive acts or practices, as well as unfair methods of competition, as well as a host of other
related statutes. The FTC’s economy-wide mission has, over its century long history, adapted to
the development and deployment of new technologies, many of which pose novel and important
challenges to the consumers, workers, and honest businesses who depend on our markets being
free and fair. AI is the latest of such challenges.
AI-enabled systems are being deployed across a wide range of industries—including
entertainment, healthcare, finance, advertising, and transportation. Although policymakers have
debated the disruptive potential of AI for decades, the pace of the technology’s development and
rollout has accelerated in recent years, and many generative AI tools have been made available to
companies and to the public. For example, some of the largest companies in the world have used
machine learning to process user inputs, identify fraud, and target advertising, while AI chatbots
and other applications have publicly launched and reached hundreds of millions of people. As AI
increasingly becomes a larger part of daily life, the technology could transform the way we live,
work, and interact.
The manner in which companies are developing and releasing generative AI tools and
other AI products, however, raises concerns about potential harm to consumers, workers, and
small businesses.1 The FTC has been exploring the risks associated with AI use, including
1 See, e.g., Lina M. Khan, We Must Regulate A.I. Here’s How, N.Y. Times (May 6, 2023),
https://www.nytimes.com/2023/05/03/opinion/ai-lina-khan-ftc-technology.html; Fed. Trade Comm’n, Combatting
Online Harms Through Innovation, Rep. to Cong, at 9-37, 56, (June 16, 2022),
https://www.ftc.gov/system/files/ftc_gov/pdf/Combatting%20Online%20Harms%20Through%20Innovation%3B%
20Federal%20Trade%20Commission%20Report%20to%20Congress.pdf.

FTC Comment
October 30, 2023
3
violations of consumers’ privacy,2 automation of discrimination and bias,3 and turbocharging of
deceptive practices,4 imposter schemes,5 and other types of scams.6
From an enforcement perspective, the FTC has been using its existing legal authorities to
take action against illegal practices involving AI. For instance, the FTC alleged that Amazon and
Ring used highly private data—voice recordings collected by Amazon’s Alexa voice assistant
and videos collected by Ring’s internet-connected home security cameras—to train their
algorithms while violating customers’ privacy.7 The Alexa matter, in particular, underscored that
the Children’s Online Privacy Protection Act Rule’s prohibition on the indefinite retention of
children’s data and similar legal rules are not superseded by claims from businesses that data
must be indefinitely retained to improve machine learning algorithms.8 In recent months, the
2 See Elisa Jillson, Hey, Alexa! What are you doing with my data?, Fed. Trade Comm’n (June 13, 2023),
https://www.ftc.gov/business-guidance/blog/2023/06/hey-alexa-what-are-you-doing-my-data.
3 See Elisa Jillson, Aiming for truth, fairness, and equity in your company’s use of AI, Fed. Trade Comm’n, (Apr. 19,
2021), https://www.ftc.gov/business-guidance/blog/2021/04/aiming-truth-fairness-equity-your-companys-use-ai; see
also Rohit Chopra, Dir., Consumer Financial Protection Bureau, et al., Joint Statement on Enforcement Efforts
Against Discrimination and Bias in Automated Systems (Apr. 25, 2023),
https://www.ftc.gov/system/files/ftc_gov/pdf/EEOC-CRT-FTC-CFPB-AI-Joint-Statement%28final%29.pdf.
4 See Michael Atleson, Keep your AI claims in check, Fed. Trade Comm’n (Feb. 27, 2023),
https://www.ftc.gov/business-guidance/blog/2023/02/keep-your-ai-claims-check; Michael Atleson, The Luring Test:
AI and the engineering of consumer trust, Fed. Trade Comm’n (May 1, 2023), https://www.ftc.gov/business-
guidance/blog/2023/05/luring-test-ai-engineering-consumer-trust; Michael Atleson, Watching the detectives:
Suspicious marketing claims for tools that spot AI-generated content, Fed. Trade Comm’n (July 6, 2023),
https://www.ftc.gov/business-guidance/blog/2023/07/watching-detectives-suspicious-marketing-claims-tools-spot-
ai-generated-content; Michael Atleson, Can’t lose what you never had: Claims about digital ownership and creation
in the age of generative AI, Fed. Trade Comm’n (Aug. 16, 2023), https://www.ftc.gov/business-
guidance/blog/2023/08/cant-lose-what-you-never-had-claims-about-digital-ownership-creation-age-generative-ai;
Lesley Fair, For business opportunity sellers, FTC says “AI” stands for “allegedly inaccurate,” Fed. Trade
Comm’n (Aug. 22, 2023), https://www.ftc.gov/business-guidance/blog/2023/08/business-opportunity-sellers-ftc-
says-ai-stands-allegedly-inaccurate.
5 In 2020, long before the public release of large language models, the FTC held a workshop highlighting the risks
of voice cloning technology that could supercharge imposter schemes. See Fed. Trade Comm’n, You Don’t Say: An
FTC Workshop on Voice Cloning Technologies (Jan. 28, 2020), https://www.ftc.gov/news-
events/events/2020/01/you-dont-say-ftc-workshop-voice-cloning-technologies. Since then, the FTC has proposed a
trade regulation rule entitled “Rule on Impersonation of Government and Businesses,” which would prohibit the
impersonation of government, businesses, or their officials, including by the use of generative AI. See 87 Fed. Reg.
62741 (Oct. 17, 2022).
6 See Michael Atleson, Chatbots, deepfakes, and voice clones: AI deception for sale, Fed. Trade Comm’n (Mar. 20,
2023), https://www.ftc.gov/business-guidance/blog/2023/03/chatbots-deepfakes-voice-clones-ai-deception-sale.
7 Press Release, Fed. Trade Comm’n, FTC and DOJ Charge Amazon with Violating Children’s Privacy Law by
Keeping Kids’ Alexa Voice Recordings Forever and Undermining Parents’ Deletion Requests (May 31, 2023),
https://www.ftc.gov/news-events/news/press-releases/2023/05/ftc-doj-charge-amazon-violating-childrens-privacy-
law-keeping-kids-alexa-voice-recordings-forever; Press Release, Fed. Trade Comm’n, FTC Says Ring Employees
Illegally Surveilled Customers, Failed to Stop Hackers from Taking Control of Users’ Cameras (May 31, 2023),
https://www.ftc.gov/news-events/news/press-releases/2023/05/ftc-says-ring-employees-illegally-surveilled-
customers-failed-stop-hackers-taking-control-users.
8 See Statement of Commissioner Alvaro M. Bedoya, Joined by Chair Lina M. Khan and Commissioner Rebecca
Kelly Slaughter, In the Matter of Amazon Alexa (United States v. Amazon.com, Inc.), at 1 (May 31, 2023),

FTC Comment
October 30, 2023
4
FTC secured a temporary restraining order against a business-opportunity seller that claimed to
use AI to make clients profitable and successful.9 The FTC has also made clear that a business
that relies on algorithmic decision-making must ensure that the algorithm is not resulting in
unlawful bias.10 Furthermore, the FTC charged WealthPress with using deceptive claims to sell
consumers investment-advising services—often claiming that the services’ recommendations
were based on an algorithm created by a purported expert.11
The rapid development and deployment of AI also poses potential risks to competition.
The rising importance of AI to the economy may further lock in the market dominance of large
incumbent technology firms.12 These powerful, vertically integrated incumbents control many of
the inputs necessary for the effective development and deployment of AI tools, including cloud-
based or local computing power13 and access to large stores of training data. These dominant
technology companies may have the incentive to use their control over these inputs to unlawfully
entrench their market positions in AI and related markets, including digital content markets. In
addition, AI tools can be used to facilitate collusive behavior that unfairly inflates prices,
precisely target price discrimination, or otherwise manipulate outputs.14 The FTC is empowered
under Section 5 of the FTC Act to protect the public against unfair methods of competition,
including when powerful firms unfairly use AI technologies in a manner that tends to harm
competitive conditions.15 Enforcers must be vigilant to protect the public from the potential
https://www.ftc.gov/system/files/ftc_gov/pdf/Bedoya-Statement-on-Alexa-Joined-by-LK-and-RKS-Final-
1233pm.pdf.
9 Press Release, Fed. Trade Comm’n, FTC Action Stops Business Opportunity Scheme That Promised Its AI-Boosted
Tools Would Power High Earnings Through Online Stores (Aug. 22, 2023), https://www.ftc.gov/news-
events/news/press-releases/2023/08/ftc-action-stops-business-opportunity-scheme-promised-its-ai-boosted-tools-
would-power-high-earnings.
10 Press Release, Fed. Trade Comm’n, FTC Chair Khan and Officials from DOJ, CFPB and EEOC Release Joint
Statement on AI (Apr. 25, 2023), https://www.ftc.gov/news-events/news/press-releases/2023/04/ftc-chair-khan-
officials-doj-cfpb-eeoc-release-joint-statement-ai.
11 Press Release, Fed. Trade Comm’n, FTC Suit Requires Investment Advice Company WealthPress to Pay $1.7
Million for Deceiving Consumers (Jan. 13, 2023), https://www.ftc.gov/news-events/news/press-releases/2023/01/ftc-
suit-requires-investment-advice-company-wealthpress-pay-17-million-deceiving-consumers.
12 See Staff in the Bureau of Competition & Office of Technology, Generative AI Raises Competition Concerns,
Fed. Trade Comm’n (June 29, 2023), https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/2023/06/generative-
ai-raises-competition-concerns.
13 The FTC has taken enforcement action to protect competition and innovation in markets for computer processors
with potential AI applications. See Complaint, In the Matter of Nvidia/Arm, Docket No. 9404, Comm’n File No.
2110015 (Dec. 2, 2021),
https://www.ftc.gov/system/files/documents/cases/d09404_part_3_complaint_public_version.pdf.
14 See, e.g., Khan, supra, note 1 (discussing algorithmic price collusion and price discrimination); see also Huskey v.
State Farm Fire & Cas. Co., No. 22 C 7014, 2023 WL 5848164, at *9 (N.D. Ill. Sept. 11, 2023) (discussing
discriminatory insurance claim processing algorithm).
15 See Fed. Trade Comm’n, Policy Statement Regarding the Scope of Unfair Methods of Competition Under Section
5 of the Federal Trade Commission Act, Comm’n File No. P221202 (Nov. 10, 2022),
https://www.ftc.gov/system/files/ftc_gov/pdf/P221202Section5PolicyStatement.pdf.

FTC Comment
October 30, 2023
5
dangers of novel AI technologies while helping to foster competitive and innovative AI markets
that benefit the economy and consumers.
Although AI-based technology development is moving swiftly, the FTC has decades of
experience applying its authority to new and rapidly developing technologies. Vigorously
enforcing the laws over which the FTC has enforcement authority in AI-related markets will be
critical to fostering competition and protecting developers and users of AI, as well as people
affected by its use. Firms must not engage in deceptive or unfair acts or practices, unfair methods
of competition, or other unlawful conduct that harms the public, stifles competition, or
undermines the potentially far-reaching benefits of this transformative technology. As we
encounter new mechanisms of violating the law, we will not hesitate to use the tools we have to
protect the public.
III. Copyrights and AI-generated Content
The FTC has an interest in many of the difficult questions with which the Copyright
Office has been grappling about where to draw the line between human creation and AI-
generated content. For instance, not only may creators’ ability to compete be unfairly harmed,
but consumers may be deceived when authorship does not align with consumer expectations,
such as when a consumer thinks a work has been created by a particular musician or other artist,
but it has been generated by someone else using an AI tool.
The FTC is also concerned with questions surrounding liability issues arising from the
development or deployment of generative AI. How should liability principles apply to harm
caused by AI tools trained on creative work that are used to generate new content? How should
liability be apportioned among users, developers of AI tools, and the developers of the training
dataset? How should this analysis take into account that training data is often scraped from
sources hosting pirated content? Should the availability of disclosures regarding training
materials or the lack of such disclosures affect liability? How would such liability principles
apply to open-source AI models? These liability questions implicate consumer protection and
competition policy. For instance, under certain circumstances, the use of pirated or misuse of
copyrighted materials could be an unfair practice or unfair method of competition under Section
5 of the FTC Act.
The use of AI technology raises significant competition and consumer protection issues
beyond questions about the scope of rights and the extent of liability under the copyright laws.
As the courts apply the doctrine of fair use to the training and use of AI,16 the evolution of the
doctrine could influence the competitive dynamics of the markets for AI tools and for products
with which the outputs of those tools may compete. Conduct that may violate the copyright
laws––such as training an AI tool on protected expression without the creator’s consent or selling
output generated from such an AI tool, including by mimicking the creator’s writing style, vocal
or instrumental performance, or likeness—may also constitute an unfair method of competition
or an unfair or deceptive practice, especially when the copyright violation deceives consumers,
16 See. e.g., Thomson Reuters Enter. Ctr. GmbH v. Ross Intel. Inc., No. 1:20-CV-613-SB, 2023 WL 6210901, at *6-
11 (D. Del. Sept. 25, 2023) (fair use defense in a case involving scraping content for training a generative AI tool).

FTC Comment
October 30, 2023
6
exploits a creator’s reputation or diminishes the value of her existing or future works, reveals
private information, or otherwise causes substantial injury to consumers. In addition, conduct
that may be consistent with the copyright laws nevertheless may violate Section 5.17 Many large
technology firms possess vast financial resources that enable them to indemnify the users of their
generative AI tools or obtain exclusive licenses to copyrighted (or otherwise proprietary) training
data, potentially further entrenching the market power of these dominant firms.18 These types of
issues not only touch on copyright law and policy but also implicate consumer protection and
competition concerns across a wide range of industries.19
IV. FTC Roundtable on the Creative Economy and Generative AI
To assist the Copyright Office in assessing the harms generative AI systems pose to
creative professionals, the FTC has appended to this Comment the transcript of its October 4,
2023, “Creative Economy and Generative AI” roundtable. The roundtable explored how the
development and deployment of AI tools that generate text, images, and audio is impacting open
and fair competition. The session included musicians, authors, actors, artists, software
developers, and other creative professionals who discussed the ways that emerging AI tools are
reshaping each of their respective industries and how they are responding to the development and
use of AI to generate new content. Some of the views shared by participants directly implicate
the work of the FTC; other concerns are outside the scope of our agency’s authorities.
Although the participants covered a wide range of experiences with and concerns about
the transformative and disruptive nature of generative AI, a few themes emerged from the
roundtable:
1. Participants said that their work has been used to train generative AI models
without their consent. In some instances, participants said that work was taken
from sources that themselves have pirated content, circumventing copyright
protections. Scraping work from public websites without consent particularly
harms creative professionals whose businesses depend on having public portfolios
to attract customers. In the long term, fear about having creative work scraped and
used in unauthorized and potentially repugnant or displacing ways could
disincentive creators from sharing their work online. Participants were further
concerned that consent they have provided in the past in a different context will
be used to justify the uncontemplated use of their work for AI training. In
particular, creators expressed the desire to avoid having their work collected,
17 See FTC v. Real Prods. Corp., 90 F.2d 617, 619 (2d Cir. 1937) (“A copyright is not a license to engage in unfair
competition.”).
18 See, e.g., Steve Lohr, IBM Tries to Ease Customers’ Qualms About Using Generative A.I., N.Y. Times (Sept. 28,
2023), https://www.nytimes.com/2023/09/28/business/ibm-ai-data.html.
19 See, e.g., Alvaro M. Bedoya, Your Body, Your Work? Hollywood Is Scanning Actors’ Moves. What That Could
Mean for the Rest of Us, LA Times (Sept. 4, 2023), https://www.latimes.com/opinion/story/2023-09-04/writers-
strike-artificial-intelligence-actors-body-scans-chatgpt-jet-li (discussing competition concerns related to the use of
AI) (further noting that “[l]ater cases make clear that the FTC Act may prevent a powerful market participant from
forcing a weaker one to act against its interests, particularly when it significantly reduced competition”).

FTC Comment
October 30, 2023
7
used, and shared for purposes beyond the scope of the original agreement (e.g.,
data collected with consent for research purposes that is subsequently used for
commercial purposes). In addition, many creative professionals do not own the
rights to the work they create, further limiting their ability to control whether their
work is used by AI tools.
2. Participants stated that, even when some mechanisms have been implemented to
offer creators consent and control over whether their work is used in AI training,
these measures have been insufficient and ineffective. Participants described “opt-
out” frameworks as an example of such measures and said that an opt-out default
puts the burden on creators to police an ever-changing marketplace. Participants
said that they are often unable to tell whether their work has been included in AI
training, in part because many AI model developers have chosen not to disclose
what is in their training data. Even if creators were able to determine that their
work has been included in AI training, participants said there is no easy way to
remove data from a trained model without retraining. Participants expressed
desire for opt-in frameworks, where AI developers seek authorial consent and
clearly explain how they intend to use their work, ideally with appropriate credit
and compensation.
3. Participants expressed concerns about transparency and disclosure with respect to
both the data used to train AIs and the provenance of new works generated by AI
tools. On the input side, as noted above, participants said it is difficult for them to
tell whether their work has been included in AI training. Participants asked for
mandatory disclosure of the content of training data sets. Participants also
expressed desire for clear and specific disclosures about the intended uses of their
work. On the output side, AI-generated content can easily flood markets, making
it difficult for customers and other stakeholders to discern whether content is AI
generated. AI-generated content can mimic the style of specific creators, and
users of generative AI tools can exploit the name and reputation of the creator to
gain sales and potentially compete with the creator. Participants said that when
generative AI tools use artists’ faces, voices, and performances without
permission to make digital impersonations, it can not only create consumer
confusion, but it also can cause serious harm to both fans and artists.
4. Participants stated that they are not intrinsically opposed to AI and that they have
already seen how in certain instances these technologies can prove fruitful to
creative professionals. Participants said that AI-enabled tools and technology
could have the potential to assist the creative community and consumers but that
guardrails, including strong defaults to minimize data collection and ensure
consent, are desperately needed. Participants said that, when used ethically and in
a manner that recognizes intellectual property rights, AI could potentially further
opportunities for artists, including by creating access to employment for people
with disabilities and those who would otherwise be prevented from pursuing work
in the entertainment industry. For example, one participant said that he was
working on developing a synthetic version of his voice that would allow him to
perform work he would not otherwise be able to do, like read a newspaper cover-

FTC Comment
October 30, 2023
8
to-cover every morning. This participant emphasized that not only was this being
done with his consent, control, and compensation, but more generally that
generative AI tools should enhance and not replace the work of creative
professionals.
5. Participants described a great power imbalance between those whose data is being
used and those who are using the data. Certain types of creative professionals are
particularly vulnerable to exploitation involving generative AI, especially those
who lack awareness or leverage in negotiating contract terms or the resources to
enforce their rights. Participants said most creative professionals do not have
access to unions to protect their interests. Writers and other artists basically
operate as small businesses, each competing in the marketplace to sell their work.
They develop a unique style, voice, and brand to distinguish themselves.
Participants stated that generative AI fundamentally disrupts that market in ways
that could be devastating to the creative economy.
V. Conclusion
AI, in particular generative AI, is still evolving rapidly, but it already has the potential to
transform many industries and business practices. Notably, there is no AI exemption from the
laws on the books. Accordingly, the FTC will vigorously use the full range of its authorities to
protect Americans from deceptive and unfair conduct and maintain open, fair, and competitive
markets. We look forward to collaborating with the U.S. Copyright Office as competition and
consumer protection issues surrounding AI-enabled tools and technologies continue to develop.
Enclosure:
Transcript from the FTC’s October 4, 2023, roundtable on the Creative Economy and Generative
AI

Page 1 of 40
Creative Economy and Generative AI - October 4, 2023
Madeleine Varner:
Hello and welcome to the FTCs roundtable discussion on the creative economy and generative
AI. My name is Madeleine Varner and I'm a Senior Technology Advisor in the Office of
Technology with a background in investigative research. Today we'll be discussing artificial
intelligence tools that can output content on command, also known as generative AI. We've
gathered professionals from a broad range of creative fields to discuss how these tools are
reshaping their respective lines of work and how they're responding to these changes.
Participants today include screenwriters, actors, programmers, editors, musicians, models, and
more. Thank you all for being here and voicing your unique perspectives.
Before we begin, please note that the FTC is recording this event, which may be maintained,
used, and disclosed to the extent authorized or required by applicable law, regulation, or order,
and it may be made available in whole or in part, the public record, in accordance with the
commission's rules. In practice, this means that we'll be sharing a recording of this as well as a
transcript on the event webpage after this concludes.
Now, we're going to be hearing some opening remarks from Chair Khan. Chair Khan?
Chair Lina Khan:
Great, thanks so much. Hi everybody. So great to be here with you all today. The FTC is hosting
this roundtable to hear directly from creators about how generative AI is affecting your work and
livelihood, and I'm just so glad that we have this opportunity to be engaging in a very timely and
important discussion. I wanted just to provide some backdrop for what the FTC’s role is here.
The FTC was created 109 years ago against the backdrop of an industrial revolution that had
delivered enormous technological progress, but had also concentrated power and control in the
hands of a few. For example, the advent and expansion of railroads meant that farmers could
now move their wares across the country dramatically expanding the number of markets that

Page 2 of 40
they could reach, but consolidated control over the railroads also meant that farmers were often
at the mercy of a single company that had the power to arbitrarily hike rates and set
discriminatory terms.
And the anti-monopoly movement and antitrust laws that followed were fundamentally about
securing terms of fair dealing across markets to ensure that Americans be it farmers, small
businesses, workers, or consumers could engage in commerce on fair terms no matter what the
technological advances of the day were, and so Congress created the FTC to enforce these rules
of fair competition, and a key part of our mandate is ensuring that these core principles continue
to apply even as technologies and business practices evolve. Lawmakers actually explicitly gave
the FTC the authority to conduct detailed market inquiries so that we could track and keep pace
with new market developments.
Today as we see growing use of automated systems, including those sometimes marketed as
artificial intelligence, we again want to make sure that we're keeping pace, that we're fully
understanding how these new tools could be affecting people on the ground in positive ways, but
also potentially in harmful ways and potentially unlawful ways.
One particular area, of course, is generative AI, and I think we're all waiting to see what the full
impact of generative AI will be, but there's little doubt that this technology could really
transform how we live and work and communicate. And at the FTC, we're in particular looking
closely at how some of these tools could turbocharge fraud, entrench the dominance of firms that
control the necessary raw inputs like cloud services and computing power, and potentially lock
in business models that incentivize the endless surveillance of our personal data. As these
technologies involved, we're fully committed to using all of our tools and authorities to maintain
open, fair, and competitive markets without tolerating exploitative or deceptive business
practices, and we think it's extraordinarily important to not assume that there's some inevitable
endpoint about how these technologies will be used, but instead recognize that the laws and
policies against which these new tools are introduced will significantly shape who benefits and
who is harmed. That's really what we see as the FTC’s rule.
We've been very clear that we want to make sure the market understands that there's no AI
exemption to the laws on the books, so all of the laws that already prohibit unfair methods of
competition or collusion or discrimination or deception, all of those laws still entirely apply. I
know that generative AI in particular poses a unique set of opportunities and challenges to
creative industries. We've already heard significant concern about how these technologies could
virtually overnight significantly disempower creators and artists who may watch their life's
creation be appropriated to models over which they have no control. Really, really eager to be
hearing from all of you.

Page 3 of 40
Over the summer, I had a chance to meet with some of the writers who were part of the WGA
and picketing and really understand from them what their concerns were. I was really thrilled to
see that they were able to reach an agreement that includes certain protections for writers from
how AI could be deployed.
Very much recognize that this is a fast-moving dynamic situation, but it's clear that for enforcers
and regulators to be keeping pace and understanding what's happening on the ground is going to
be absolutely essential, and we really couldn't do that without you all, so thank you so much for
taking the time to share your experiences and views with us today. Our Office of Technology,
which we launched earlier this year, is really a critical part of this effort. We wanted to make
sure we have the skill sets on board to help us understand how are these tools really working and
know what's really going on. And they, in close partnership with the Bureau of Consumer
Protection, Bureau of Competition and Office of Policy Planning, were essential to putting
today's event together, so my deep gratitude to everybody who was involved with that as well.
Really looking forward to hearing from you all. I know Maddy and others have teed up some key
questions and then we can have a discussion. With that, I will pass it back to Maddy to kick it
off.
Madeleine Varner:
Thank you, Chair Khan. We're going to turn it over to Commissioner Slaughter for her remarks.
Commissioner.
Rebecca Slaughter:
Thank you, Maddy, and thanks to Chair Khan and all the FTC staff members who've worked to
make today's event possible, and thank you in particular, to all of our roundtable participants,
artists from such a wide range of creative media. Today's discussion involves two very different
concepts, art and technology that are connected by an essential input, humans. Art is
fundamentally human. Humans may use technology to assist in creating art, but something
cannot be art without human input. Technology is, by definition, not human. Yet technology
including generative artificial intelligence requires human intelligence. While humans may
endeavor to make generative AI that is ever more intelligent, it cannot and will not replace
human creativity. The value of creative arts to society is so fundamental that it is enshrined in the
Constitution. You all are the humans who have mastered a craft and you share it with all of us for
the benefit of the public and society as a whole.

Page 4 of 40
In return, such works of art may be granted copyright protection. Copyright provides your
livelihood and the ability to continue to create and promote further creation and learning.
Generative AI poses important questions and concerns about how copyright law and policy must
be applied or adapted to continue to both protect creators and promote the useful arts, but
copyright is not and cannot be the only tool to address the deeply personal concerns creators hold
about how their works are used. The Writer's Guild of America has demonstrated the power of
collective bargaining to secure important rights on how they will interact with, use, and be
subjected to generative AI. Many but not all states have laws that provide rights to publicity,
which may provide avenues for legal protection and compensation, and as the chair noted, the
FTCs prohibitions against unfair and deceptive practices and unfair methods of competition
apply to applications of AI just as much as they have to every other new technology that's been
introduced in the market over the last hundred years.
These are powerful tools we can use on behalf of creators, workers, and consumers, and there
may also be gaps in the law that need to be filled. As artificial intelligence processes, uses, and
applications evolve, we cannot lose sight of the fundamental truth that technology is a tool to be
used by humans. Humans are not and should not be used by technology. One of my favorite
things about the work of the FTC is how it matters to real people in their real everyday lives,
how it matters for real humans. I return to where I started, that art is fundamentally human. As
the FTC considers how its tools can help fight fraud, deception, and unfair methods of
competition related degenerative AI, it's critical to hear from stakeholders and that's why I'm so
grateful to all of our speakers for their time and insight today.
Thanks again and I'll pass it back to you, Maddy.
Madeleine Varner:
Thank you, Commissioner Slaughter, and just to note, my remarks today don't necessarily reflect
the views of the commission or any individual commissioner.
The Office of Technology is fortunate to be to be able to organize today’s roundtable alongside
staff from across the agency—I want to give an immense thank you to those in the Bureau of
Consumer Protection, Bureau of Competition, and Office of Policy Planning for your efforts
leading to today.
We look forward to hearing the lived perspectives of those from across different creative
industries, ones that are actively being transformed by the development of new generative AI
systems.

Page 5 of 40
Our Office works to keep pace with emerging developments in digital markets, including in the
rapidly involving generative AI space. Our software engineers, researchers, and practitioners use
their deep expertise to examine the different layers of these technologies–including training data
and infrastructure used to develop AI models and the models themselves in order to better
understand how these technologies are built and the ways it might impact market participants and
pose harms to consumers, workers, and small businesses.
In addition to understanding the technical underpinnings of generative AI, we root our internal
expertise in the day-to-day lived experience of those who are most impacted. Shining a light on
how emerging technologies impact people and communities is an important way for us to orient
our law enforcement and policy work.
We know that generative AI relies on a critical input to function, large, diverse data sets of
human content. One method to build such data sets is web scraping, often performed unseen, and
without the knowledge of creators whose work is being collected. We have heard from
individuals, some of which are participants today, that this dynamic deeply impacts those who
share work online, particularly creative professionals whose livelihoods can depend on having a
public portfolio or presence to attract customers. Today, we'll have the opportunity to hear
directly from these people.
Today's roundtable is an important opportunity to ensure that we are assessing the full range of
the impacts of generative AI on creative communities and help ensure that we are using our full
set of authorities to tackle unfair or deceptive acts and practices and unfair methods of
competition in these fields.
We're grateful to the creative professionals who have lent their time today to share their
experiences, and we're looking forward to an engaging and insightful roundtable.
And with that, we're going to kick things off with Duncan Crabtree-Ireland, the National
Executive Director and Chief Negotiator for SAG-AFTRA.
Duncan Crabtree-Ireland:
Thanks so much, Madeleine. Can you all hear me okay? I'm hoping so. Hi everybody. I'm
Duncan Crabtree-Ireland. I'm the National Executive Director and Chief Negotiator of SAG-
AFTRA, and we are the union that represents over 160,000 members who are the faces and

Page 6 of 40
voices that entertain and inform the world, and I just want to say thank you for the opportunity to
speak on a topic that's so important, not just for creative talent, but for workers across all
industries.
As AI technology, and generative AI in particular, affects anyone who's concerned with consent
and with protecting their own intellectual property rights. Generative AI, generative artificial
intelligence, poses a threat to the livelihoods of many. But to be clear, we at SAG-AFTRA are
not opposed to new technologies and we're not opposed to the existence or even the use of AI. AI
in particular is an essential tool that is going to be widely used in the decades to come and it can
have a variety of creative and innovative uses in the entertainment industry, and the
implementation of AI we are looking to achieve would result in the technology augmenting
human creativity rather than replacing it.
When used ethically and in a manner that recognizes intellectual property rights, AI can help
people in their careers and can further opportunities. It can create access to employment for
people with disabilities and those who would otherwise be prevented from pursuing work in the
entertainment industry. What SAG-AFTRA is eager to do is to channel the benefits of AI into a
future that's beneficial to our members, to workers in other industries, and to the public in
general. The key is that the companies using AI technology must be required to get the informed
consent of any individuals whose voice, likeness, performance, persona, or intellectual property
is being used to generate content and companies need to compensate these individuals fairly.
Informed consent and compensation addresses a lot of the most important ethical questions
attached to how generative AI works. It also would ensure that the hundreds of thousands of
individuals who work in the entertainment industry will be paid for the part they play in creating
content for these companies who are using human beings to create their profit.
It's important to understand that all AI generated content originates from a human creative
source. No AI algorithm is able to make something out of nothing, and that human generated
content that's used in the training data reflects real and substantial work and its intellectual
property and it deserves legal protection. There's a bit of a double standard that can be seen
happening in a lot of these conversations around AI. After all, if an individual decided to infringe
on one of these company's copyright protected content and distribute it without paying for the
licensing rights, that individual would face a great deal of financial and legal ramifications.
So, why is the reverse not true? Shouldn't the individuals whose intellectual property was used to
train the AI algorithm be at least equally protected? An actor's brand is their voice, as is their
likeness and their unique persona, and no company should be able to appropriate that and use it

Page 7 of 40
however they wish without permission. What we are proposing is about keeping our world and
our industry human-centered. AI and its algorithms must be here to serve us, not the other way
around. Thank you.
Madeleine Varner:
Thank you so much. Next, we're going to hear from John August, a screenwriter and member of
the Writer's Guild of America West Negotiating Committee. John?
John August:
Thank you. It's a pleasure to be here.
I'm here on behalf of the Writers Guild of America West, a labor union representing thousands of
writers in film, television, and streaming series. Our members and the members of Writers Field
of America East have just concluded a 148-day strike where artificial intelligence was a key
issue. As Duncan just noted, our fellow artists at SAG-AFTRA are still on strike, with AI as a
core issue for them as well, but the fight for protection over our craft and livelihoods doesn't stop
at the bargaining table. While we have been able to achieve groundbreaking protection for
writers, we need public policy solutions, too.
Obviously, copyright is an area of government scrutiny, both the copyright ability of AI
generated work and the degree to which training AI models infringes upon copyright. WGA
writers do not hold copyright to most of the scripts we write; those are works made for hire, so
studios— our employers— hold the copyright, but through the power of our union, we have over
the decades negotiated an assortment of contractual rights in the works we create, including the
right to payment for reuse of our work. It's a good reminder that while copyright is important, it's
not the end of the story when it comes to protecting artists.
The Guild’s new agreement offers helpful guidance in thinking about future public policy on AI.
Our agreement defines that AI is not a writer and the material it generates is not equivalent to
human writing for purposes of our contract. That means that AI cannot rewrite us, nor can it
compete with a human writer for credit and the associated financial benefit of that credit.
Further, the studios now have to tell us if they are providing us with material generated by AI
and then it cannot require us to use AI tools.

Page 8 of 40
We won these protections because we're a strong union that successfully carried off a nearly five
month strike. But we need to remember that most writers and most artists in this country don't
have unions to protect them. It's best to think of writers and other artists as tiny businesses, each
competing in the marketplace to sell their work. Writers and artists each develop a unique style,
voice and brand in order to distinguish themselves. AI fundamentally disrupts that market in
ways that could be devastating to the creative economy.
Large language models like the one that underpins ChatGPT have scraped massive volumes of
data, including our words and our unique perspectives. This is theft, not fair use. Our works—
protected by copyright and our own contractual rights—are being used entirely without our
authorization, without any attribution or compensation. Right now, you could ask ChatGPT to
write something “in the style of” a particular writer, and they would try to do that—appropriate
the unique voice of a writer without that writer's consent.
As FTC Commissioner Alvaro Bedoya recently suggested, this could constitute an unfair method
of competition. It is using stolen goods to undercut the price of a seller and create market
confusion, and it's not a hypothetical. Right now, authors are finding AI generated knock-offs of
their work published on Amazon. They're having to fight to get those fakes taken down, and
protect their brands.
This form of AI appropriation may also have consumer implications. From electronics to organic
eggs, consumers expect to be told the origin of a product and its authenticity. Consumers make
choices based on that information. The same will likely be true with AI.
So, with our strike and this contract, the Writers Guild was able to win groundbreaking AI
protections for writers. But it's important to remember that our deal only covers the film and
television studios. Most of the real work in AI is being done by companies like Google,
Facebook, and OpenAI with which the Guild has no contractual relationship. Public policy will
play a crucial role in protecting our members.
In conclusion, AI does have the potential to assist the creative community, but only with the
consent of that community. Without guardrails, AI poses a profound threat to writers and the
integrity of our creative work. AI and its use raises major questions of intellectual property,
transparency of competition, all of which require careful oversight to protect the creative
economy.

Page 9 of 40
Thank you for the opportunity to speak on behalf of film and television writers.
Madeleine Varner:
Thank you. Next we're going to hear from Neil Clarke, Founder and Editor of the Award-
Winning Sci-Fi magazine, Clarkesworld. Neil?
Neil Clarke:
Hi, I'm the Publisher and Editor of Clarkesworld, a science fiction and fantasy magazine that has
been publishing for 17 years. We publish stories not only from established writers, but also new
voices from all over the world. These people represent the future of my field, and it is essential to
maintain the avenues through which they can be discovered.
Like many of my colleagues, we do this by maintaining an open submissions process in which
anyone can submit their stories for consideration. Not long after ChatGPT was released, we
started noticing some unusual submissions in our queue and quickly realized that they were
generated. It started small. A few in November, around 50 in December, over a hundred in
January. In the first 20 days of February, it spiked and we received over 500. On the morning of
the 20th alone, we received over 50 and the daily trends were indicating that we would double
our normal monthly submission volume of 1100 by the end of the month.
This was unsustainable, so for the first time in over a decade, we closed submissions for
something other than a software update. We needed breathing room to process what we had
received and time to figure out some way to block, deter, or minimize these submissions. Even
before this happened, we had a standing no AI policy. We knew these models were trained
without permission on the copyrighted works of others, and it didn't sit right with us. It's also
unclear that the people submitting these works actually own the rights they require. There are
those who have argued that we should be considering these stories on their own merits rather
than rejecting them for being generated. I can confidently state that they were among the worst
stories we've ever received.
The problem is not quality at the moment, but rather the sheer speed and volume at which these
works can be produced. The best way to describe them is to call them spam, but it's spam at a
level we've never seen before. The generated submissions came from outside the science fiction
fantasy community, typically from countries with a much lower cost of living and higher
unemployment. We traced the source of much of this activity to YouTube, TikTok, and blog
content hosted by side hustle con artists that would waive a stack of cash and make false claims

Page 10 of 40
about earning riches from ChatGPT. Their intent was to earn advertising revenue or sell classes
to those that tried and failed. They knew these tactics would not get people published. We
happened to be on a list of a hundred or so magazines that were used as part of these schemes.
Over the months that followed, copycats would refine their approach.
Several creating videos that specifically targeted Clarkesworld and even added directions on how
to avoid detection. We reached out to YouTube and others in the hopes of having these videos
taken down, but no one ever responded. I'm also the developer of the submission system that's
used by Clarkesworld and a few other magazines. During the period we were closed, I spoke to
professionals with experiencing credit card fraud prevention, network security, spam detection,
and even some sympathetic people working on AI research. I evaluated many of the public and
commercial detection tools and found their claims significantly overstated. The number of false
positives and false negatives made them unusable. To make matters worse, I observed that
foreign authors were far more likely to be incorrectly flagged as AI and an experienced eye
remains the only reliable method of detection. When we reopened, the tools and techniques we
adopted worked for about two months before the spans changed tactics and the levels shot up
once again.
It's been back and forth since. Some days are good, many are bad. Unfortunately, we still have to
review each of these suspicious submissions manually, but for now, we've been able to stay
open. Between the time spent reviewing submissions and maintaining the software to control
them, my workload has easily doubled. It's been exhausting, and this is only the early days.
Submissions are still being submitted manually by people. Once they start employing bots to
submit these words, what we currently receive in a month may arrive in hours. Quantity and the
lack of reliable detection tools will eventually break our ability to do our job. This is not the limit
of my troubles with AI. Stories and art posted to our website have been utilized the training data
for language models without our knowledge or permission. Anthologies I've edited and
published have been pirated and included in the Books3 database used to train various models.
This has concerned our authors and some are even withdrawing their work as we can no longer
protect them. While it is likely that there are many positive uses for this technology, I don't
believe that authors, artists, translators, narrators, and others should be required to sacrifice their
work to improve or build these models. Progress can still be made responsibly without stepping
on creatives. Regulation of this industry is needed sooner than later, and each moment they are
allowed to continue their current practices only causes more harm. Their actions to date
demonstrate that they cannot be trusted to do it themselves. Thank you.
Madeleine Varner:

Page 11 of 40
Thank you. Next, we're going to hand it off to Bradley Kuhn, policy fellow at the Software
Freedom Conservancy. Bradley?
Bradley Kuhn:
First, I thank the FTC for organizing this panel. It's admittedly humbling to be here among these
key individuals from such a broad range of important creative endeavors. Folks will surely notice
that I'm not appearing by video today, and I, again, thank the FTC for providing a method for me
to join you today without requiring that I agree to Zoom's proprietary terms and conditions.
As a matter of principle, I avoid using any proprietary software, but in this case, it is not merely
esoteric principle. Zoom is among the many Big Tech companies that have sought to cajole users
into consent for use of their user data as training input for machine learning systems.
If consumers take anything away from my comments today, I hope they remember to carefully
read the terms and conditions of all software platforms they use, as they may have already agreed
for their own creative works to become part of the company's machine learning data sets.
I admit it may take you a week to read all of those terms, but it's sadly the only way you'll know
what rights you've inadvertently given away to Big Tech.
The creative works that I focus on, however, is the source code of software itself. Software is
unique among creative endeavors because it is so easy to separate the work that's created by
humans, which is the source code, from the form of the work that's enjoyed day to day by
consumers, which is the compiled binary. I'm an activist in the area of software freedom and
rights specifically because I believe every consumer deserves the right to examine how their
software works to modify, to improve, and to change it, be it altruistically or commercially. Free
and Open Source Software, abbreviated FOSS, aims to create through licensing and other means
an equal field for all software professionals and hobbyists alike and to grant rights to consumers
so they have true control of their own tools.
For 30 years, our community has created FOSS and made it publicly available. Big Tech for its
part continues to refuse to share most of its software in the same way. So as it turns out, nearly
all the publicly available source code in the world today is FOSS, and most is licensed under
terms that are what we call copy left, a requirement that anyone who further improves or
modifies the work must give similar permissions to its downstream users. This situation led
FOSS to become a canary in the coal mine of Big Tech's push for machine learning.

Page 12 of 40
Hypocritically, we've seen Big Tech gladly train their machine learning models with our publicly
available FOSS, but not with their own proprietary source code. Big Tech happily exploits
FOSS, but they believe they found a new way to ignore the key requirements that FOSS licenses
dictate. It's clear Big Tech ignores any rules that stand in the way of their profits.
Meanwhile, Big Tech has launched a campaign to manufacture consent about these systems. Big
Tech claims that the rules, licensing, and legislation that is applied to creative works since the
1800s in the United States are suddenly moot simply because machine learning is, in their view,
too important to be bogged down by the licensing choices of human creators of works.
In the FOSS community, we see this policy coup happening on every level from propaganda to
consumers to policy papers to even law journal articles. I realize that I sound rather pessimistic
about outcomes here. I'm nevertheless hopeful sitting here in this panel today because I see that
so many of my colleagues in other fields are similarly skeptical about Big Tech's self-serving
rhetoric in this regard. And I hope that we can work together to counter that rhetoric fully. Thank
you.
Madeleine Varner:
Thank you. Next, we're going to hear from Umair Kazi, director of Policy and Advocacy at
Authors Guild. Umair?
Umair Kazi:
Thank you, Maddy. Thank you, Chair Khan, Commissioner Bedoya, and Commissioner
Slaughter for centering the voices of creators in this important debate around generative AI. And
thank you to the FTC staff for organizing this excellent program.
I'm here today on behalf of the Authors Guild, the nation's oldest and largest professional
organization of published authors. Since its founding, the Guild has served as the collective voice
of American authors. Our 14,000 plus members include traditionally published and independent
authors, novelists in all genres and categories, nonfiction writers, journalists, historians, poets,
and translators. The Guild's core mission is to support working writers by protecting free
expression and copyright, and fighting for fair contracts and for authors' ability to earn a livable
wage.

Page 13 of 40
Generative AI poses a serious threat to the writing profession, and we believe that guardrails
around its development and use are urgently needed. AI developers have copied millions of
copyrighted works without permission. These works are not only copied many times in the
course of compiling training data sets and ingestion, but are embedded in the very fabric of the
language models. These works are also used when users prompt the AI system. It is inherently
unfair to use copyrighted works to create highly profitable tech, which is also able to produce
competing derivative works without the creator's consent, compensation, or credit.
There's a serious risk of market dilution from machine generated books and other works that can
be cheaply mass-produced, and which will inevitably lower the economic and artistic value of
human created works. We are already seeing that AI is being used to generate low quality
eBooks, impersonating authors, and displacing human authored books in the marketplaces like
Amazon. In one instance earlier this year, AI-generated books started dominating Amazon's
bestseller list in the young adult romance category. We have seen AI-generated books pop up for
sale on closely related or very similar topics as those authors have listed for pre-orders, a growth
in the number of unauthorized summaries of books, and fake books using names of renowned
authors.
Generative AI is being used to create unauthorized derivative works such as a developer using
ChatGPT to write the concluding books in George R. R. Martin's, “A Song of Ice and Fire”
Series and chatbots like the Dan Brown Chatbot. Beyond the obvious economic impact are the
more personal and painful incursions upon authors' voices and personhood. Pulitzer Prize
winning author and Authors Guild councilmember Min Jin Lee has likened the experience of
having her work used by AI to identity theft. "AI company stole my work, time, and creativity,"
she recently wrote on X/Twitter. "They stole my stories. They stole a part of me." It's not just
well-known and bestselling authors that are suffering. Freelance journalists and professional
writers of web and marketing content are reporting losing work at an alarming rate. An Authors
Guild member who writes marketing and web content reported losing 75% of their work as a
result of clients switching to AI.
A content writer featured in a piece about ChatGPT's impact in the Washington Post stated that
he had lost a half of his annual income. There's a widespread and real concern among writers that
AI will decimate the profession. In our recent survey, 69% of authors said that generative AI
threatens their careers, and an overwhelming 90% said that authors should be compensated if
their works are used in training. These concerns are born out of the experience of enduring long-
term precarity.

Page 14 of 40
As our most recent income survey found, the median writing related income for full-time authors
is just over $20,000 per year. The consequences of this precarity with generative AI is quickly
compounding reach beyond the writing community. Do we really want a world where our books
and literature are algorithmically synthesized mimicries of the richness of human experience? To
safeguard the incentives for creators to continue creating incentives so vital to our democratic
culture that they are inscribed in the Constitution, the Authors Guild is lobbying for laws,
regulations, and policies that recognize the following and require:
1. Consent and compensation. Require all generative AI companies to seek permission for
the use of creative works and to fairly compensate creators.
2. Credit and transparency. Create obligations for all AI companies to disclose what data
sets and works they use to train the systems.
3. Permission and payment for use in outputs. Require all AI companies to seek permission
and pay compensation when creative works are used in outputs or when names or
identities or titles of works are used in prompts.
4. Labeling AI-generated content.
5. No copyright for AI-generated outputs. We oppose efforts to deem AI-generated content
protectable under copyright law or through creation of even a limited suite generous
right.
Providing copyright or similar incentives to use AI to generate content will exacerbate the threat
of AI-generated content flooding and overwhelming market for human works. Thank you for
your time, and I will now turn it over back to Maddy.
Madeleine Varner:
Thank you. Next, we're going to hear from Douglas Preston, a bestselling author and a former
president of Authors Guild. Douglas?
Douglas Preston:
Thank you. All right, I'm glad to be here. Thank you, everyone.
In addition to being an author and the former president of the Authors Guild, I'm also the
plaintiff in a class action lawsuit against OpenAI, along with 15 other authors and the Guild
itself. And we're asking for damages for unauthorized use of our copyrighted work and training
and building ChatGPT. And going forward, we're asking that OpenAI and other AI developers
get permission from authors, properly license our books, and compensate us fairly for that use.

Page 15 of 40
When ChatGPT first came out, like many authors, I was fascinated with it and began playing
around with it. And at one point, I asked it to write a poem in heroic couplets about one of my
characters. And I was floored at the level of detail it knew when it generated this poem, and that's
when I realized it must've ingested many of my books.
And at the same time, many authors were discovering that ChatGPT-3 knew everything about
their books as well. And some realized it was even being used to create works that imitated their
own. My friend George R.R. Martin, who was already mentioned, was very disturbed when AI
was used to write the last book in his Game of Thrones series using his characters, his plot lines,
his settings—even his voice.
So we at the Authors Guild investigated, and here are some of the facts we uncovered. ChatGPT-
3 used more than 150,000 copyrighted books to feed into its AI system, which led us to the next
question. "Where did OpenAI get our books?" They're not just sitting out there on the web
unprotected. Well, the Authors Guild found that OpenAI got many, if not most, of these books
from pirate websites such as LibGen run out of Russia.
So these pirates had stolen our books, and then OpenAI stole them from the pirates, I guess.
ChatGPT would be lame and useless without our books. Just imagine what it would be like if it
was only trained on text scraped from web blogs, opinion screeds, cat stories, pornography and
the like.
Now, Sam Altman himself testified that books provide the really high value literary content that
large language models require. And he also testified that his goal with OpenAI is to support
creativity. But supporting creativity by stealing from creators is like claiming you're supporting
the candy store by shoplifting.
This is our life's work. We pour our hearts and our souls into our books. They aren't just
products, they're a part of us.

Page 16 of 40
And that's why we joined together, the 17 of us authors, in a class action lawsuit on behalf of all
professional novelists against OpenAI. There's nothing complicated about this lawsuit. OpenAI
illegally ingested our books to create a product that is currently valued at tens of billions of
dollars, and they did this without our consent or compensation. And as Umair mentioned, the
average full-time author in America makes only $20,000 a year. This is a classic case of Robin
Hood in reverse, stealing from the poor to give to the already obscenely rich.
In their race to be first, AI developers are swallowing everything they can get their hands on
without regard to copyright ownership, intellectual property rights, or moral rights. And they're
doing this without the slightest consideration given to supporting the livelihood of America's
creative class.
Now, it's been mentioned before. The founders of our country wrote copyright protection into the
very first clause of the Constitution—it was that important to them. They believed that their
scrappy little country one day would become the creative engine of the world, and that's what
we've become, and we can't allow AI developers to ignore copyright protection and injure the
entire literary community of our country in their mad rush to succeed. They can succeed and they
can also partner with America's authors in a mutually beneficial relationship.
Thank you.
Madeleine Varner:
Thank you. Next, we're going to hear from Tim Friedlander, president and founder of the
National Association of Voice Actors. Tim?
Tim Friedlander:
Cool. Thank you for having me here today. I am the president and co-founder of the National
Association of Voice Actors, and I'm here to represent the interests of the million strong voice
actors in the United States and the surrounding ecosystem of Americans that work with them,
ranging from engineers to script handlers to directors, producers, and PAs.
Not all voice actors are celebrities or well-known voices. Most are blue collar, working class
voice actors who are working 40 plus hours a week. Over 60% of the voice actors are located

Page 17 of 40
outside of LA and New York. We're not anti-tech or anti-AI, as many have said before. The
ability to record audio on our computer at home was game changing. I'm a child of the 80s. I still
remember the time that I died of dysentery on the Oregon Trail for the first time. We are not anti-
tech or anti-AI. We are pro voice actor.
I'm also a member of SAG-AFTRA and a professional musician for over 30 years, first in the
classical arena, and now most recently in hip hop for the last 10 years. We stand in solidarity
with our fellow creative industry artists who are in unions. But unlike them, 80% of the
voiceover industry is non-union, meaning we lack the protections and contract that organize
workers enjoy. Without the intervention of the Federal Trade Commission, AI and synthetic
voice clones will place our members at an unfair competitive disadvantage and decimate our
profession. It's difficult enough to compete against other humans. Now mom and pop voice
actors across the country have to compete against digital clones of themselves and against
multimillion and billion dollar tech companies. It's incredibly easy to use AI to capture the voice
of an actor from content available on the internet, and to use that sample to create whole works
for sale or non-commercial distribution.
Voice actors have been creating and delivering digital audio of ourselves for decades. Contracts
we signed years ago are now being used to justify the inclusion of our audio in synthetic voice
models. And every time that happens without the consent, control, and compensation of the
voice actor involved, the value of that voice actor's product, their voice and the sound of their
voice, is diluted and unfairly diminished. We consider this to be a form of theft. We agree to
license our voice to clients. The client is buying the exclusive rights to that recording, that
performance, but also the sound of our voice. We could potentially now be in conflict with a
clone of our own voice, which causes harm to both the voice actor and the company. Pepsi and
Coke can't have the same voice providing their commercials. Chevy and Ford can't have the
same voice. Currently the only protections that we have from voice actors having their audio
uploaded to these sites are the terms of service that people agree to.
Companies can change their terms of service, and currently it's easy as checking a box that says,
"I have the right to upload this audio." We've seen damages this year very clearly. A voice actor
in New York worked for a company for three years, and year four, they were let go because they
were told the company had enough of their audio, and they were going to now create a synthetic
version of their voice. In February, Twitter doxxing: Voice actors such as Michael Schwalbe and
Abbey Veffer had their voices cloned and had Twitter accounts made that gave out their personal
addresses in their own voice and said racist and homophobic things. Thousands of synthetic
voices are currently online on websites from video games all over the internet. Fan dubs are
being created and turned into adult content, and then sharing those fan dubs on YouTube using
the voices of those voice actors.

Page 18 of 40
And most recently, a voice actor in Washington state lost an audiobook job when they decided to
take the job in-house. Around that same time, the audiobook company made a press
announcement that they were now using ElevenLlabs for all of their audiobook productions,
effectively replacing all of those human narrators with synthetic voices.
Additionally, an extra burden has now been placed on the voice actors to prove that the audio
they're delivering is not AI-generated. Many voice actors have reached out to NAVA asking for
some way to prove that they delivered human-generated audio.
But it's not all terrible. I personally am working on a synthetic voice that I have consent,
compensation, and control for. There are some things that humans can't physically do, such as
narrate the New York Times cover-to-cover every morning, or provide a realistic voice for
someone who is nonverbal. But this tech should enhance and not replace voice actors.
We're asking for a few things.
First, the three C's. For us, this is consent, control, and compensation. Consent to have our voices
in the models. Control over where that model and that voice can be used. And compensation to
be paid fairly. A federal right of publicity.
If the FTC is not able to do this through rulemaking, we would request the FTC's support for
federal legislation to establish that right. Transparency of ingested content on which these
foundational models are trained in order to know if our voice is present. Protections prior to the
generation of any AI-created content that might include voices of professionals that have not
provided consent and are not being compensated. Protection of our voices' biometric data for
privacy and commercial purposes. An independent third party to verify that audio files are
ethically sourced. And finally, clear labeling of any AI-generated content to ensure the
consumers are fairly informed.
And as such, NAVA recently endorsed the [US Senators] Schatz and Kennedy’s AI Labeling Act
of 2023, which is Senate Bill 2691 that we are fully endorsing.
Thank you for your time.

Page 19 of 40
Madeleine Varner:
Thank you. We're going to hear now from Sara Ziff, founder and executive director of The
Model Alliance. Sara?
Sara Ziff:
Thanks, Maddy.
My name is Sara Ziff, and I'm the founder and executive director of the Model Alliance, a
nonprofit organization that advances workers' rights in the fashion industry. I come to this having
worked as a model myself for many years and experiencing the pitfalls of what remains a largely
unregulated industry. Models are typically hired as independent contractors through management
companies which, unlike talent agencies, are held to very few legal standards. So when we talk
about how generative AI is impacting workers, we need to consider the context of an industry
that is truly like the Wild West– where workers have few protections at baseline and also cannot
collectively bargain here in the US.
Our community of models, content creators, and other fashion creatives are concerned about the
impact of generative AI. We recently conducted an informal poll to get a better sense of their
concerns, which generally fall into two key areas:
The first is around the use of 3D body scans in connection with generative AI, and the second is
around the creation of AI-generated models – particularly AI models of color.
Increasingly, companies are asking models to undergo scans that generate a 3D model of their
body or face. In our poll, nearly 18% of models who responded had been asked to undergo a scan
by a brand or a management company.
When a model signs a management agreement, she typically hands over a power of attorney, and
thus rarely, if ever, sees her own contracts with the brands. As a result, models have little
transparency into their own business dealings.

Page 20 of 40
Those who had been scanned described not being given information about how their scans would
be used, unknowingly handing away rights to their image, and not being fairly compensated. For
people whose livelihoods are their image, this is particularly troubling in light of the rise in
deepfake technology, specifically deepfake pornography.
The second concern is around the creation of AI models and influencers, which are digitally
created, fictitious representations of human models.
Fashion workers are worried about the threat of these AI models replacing jobs – not only for
models, but also photographers, stylists, and hair and makeup artists among others.
Members in our community have expressed particular concern about companies using AI-
generated models as part of their diversity and inclusion initiatives.
For example, Shudu, a digital model who was created through AI in 2017 by the world's first all-
digital modeling agency, has appeared as a face of high-end brands such as BMW and Louis
Vuitton. Critics have called this a form of “digital blackface” since Shudu is a Black woman, and
the creator who profits off her image is a White man.
And earlier this year, Levi's announced that they are creating AI-generated models to increase
the number and diversity of their models. In an industry that has historically been discriminatory,
creating digital representations of models of various ages, ethnicities, and body types rather than
hiring and paying a diversity of real models is concerning.
I appreciate the FTC's mandate is to protect both competition and the consumer, and that the
agency is set up to prevent unfair methods of competition and deceptive practices affecting
commerce. With that in mind, there's a real risk that AI may be used to deceive investors and
consumers into believing that a company engages in fair and equitable hiring practices and is
diverse and inclusive, when they are not.

Page 21 of 40
To address these concerns, we first aim to pass the Fashion Workers Act, our signature bill,
which would establish basic labor protections for models and content creators working in New
York's fashion industry. This would help address the lack of transparency that leaves models in
the dark about how their digital image is being used, and establish a necessary foundation for
regulation around generative AI in the fashion industry.
In considering regulation, it's essential that we center the experiences and expertise of those who
will be directly impacted. And to that end, we're also developing a research study in partnership
with the Worker Institute at Cornell University to better understand the impact of generative AI
on fashion workers, particularly workers of color, and develop policy recommendations. So if
anyone is interested in learning more or getting involved, we welcome you to reach out.
At The Model Alliance, we believe now is a critical time for solidarity between workers across
creative fields who contribute heavily to our culture and economy. Unfortunately, it's not
enough to win protections through collective bargaining agreements. There are many workers,
including members of our community, who cannot engage in collective bargaining, and so we
have to ensure that they are included.
As many others have said, we're not anti-technology. We're anti-exploitation. And we believe
this round table is an important step toward ensuring our creative community is afforded the
protections we need and deserve.
Thank you for the opportunity to speak with you today.
Madeleine Varner:
Thank you. We're now going to hear from Karla Ortiz, a concept artist and illustrator.
Karla Ortiz:
Hi. FTC Chair Lina Khan, Commissioner Slaughter, Commissioner Bedoya, Maddy Varner, and
all esteemed officers at the FTC, thank you for allowing me to be a part of this amazing panel to
discuss generative AI and its impacts on the creative economy.

Page 22 of 40
My name is Karla Ortiz. I have worked as a professional concept artist, illustrator, and fine artist
for the past 14 years and have been training for that all of my life. My work has helped shape the
world's big-budget films and TV shows including Marvel Studios Loki 1 and 2, Avenger Civil
Wars, Guardian of the Galaxy 3, and most known for my design of Dr. Strange – look in the first
movie, which you can see him right there.
I deeply, deeply love what I do. Making a living as a professional requires a whole life of
practice and study. The creative economy only works when the basic tenants of consent, credit,
compensation, and transparency are followed. The country's creative economy, represented here
by notable figures from each sector of our industry, is great because it adheres to those tenant.
That success is gravely threatened by generative AI. For the first time in my life, I am worried
about my future as an artist.
Generative AI technology that uniquely could use to consume and exploit the hard work,
creativity, and innovation of others, no other prior technology is like this. I first encountered
generative AI around a year ago. I'm still shocked by its blatant exploitation. My work and the
work of almost every artist I know was stolen without consent, credit, or compensation. Without
transparency, we can't even know the extent of how much of these companies have taken. They
took our work and data to train for-profit technologies that then directly compete against us in
our own markets using generative media that is meant to mimic us.
AI companies also engage in deeply unethical practices. For example, data laundering, where a
company outsources its data collection to a third party under the pretext of research to then
immediately use that for commercial purposes. An example of this is LAION. Specifically, it's
more popular LAOIN-5B, a dataset that contains 5.8 billion text and image pairs, which again
includes the entirety of my work and the work of almost everyone I know.
Beyond intellectual property, datasets like LAOIN-5B also contain deeply concerning material,
like private medical records, non-consensual pornography, images of children. Even social media
pictures of our actual faces end up in that dataset. Many of the largest and most prolific sellers of
generative AI use these datasets. These companies describe their stolen training material as
publicly available data or openly licensed content to disguise their extensive reliance of
unauthorized use of copyrighted and private data. These companies even break their own
promises not to sell users information.
These are just few examples of the unprecedented volume of copyright infringement and
violation of other rights. No matter what these sellers say, most, perhaps all of these models rely
on stolen works. Not just this, but many of these highly profitable companies encourage users to

Page 23 of 40
use artists full names to generate imagery, exploiting our very identities and reputations. Some
artists have had their names used in props hundreds of thousands of times, maybe more. And
these numbers come from incomplete records, by the way. My own name, Karla Ortiz, has also
been used thousands of times. I never gave consent. I never got credit. I never got compensation.
Perhaps most harmfully, these exploitative products compete directly with artists and are already
replacing us. That may be the most disturbing harms of generative AI. Not vast murdering sci-fi
nonsense, but one built on works taken without credit, consent, compensation and transparency,
and marketed and used as a replacement for the creators of those works at a fraction of the cost.
The whole process is rotten.
As a side note, due to all of this, I am also a plaintiff in a class action against generative AI
image companies as well. Their plan is simple, to go as fast as possible, promising promises of
progress and innovation while normalizing the exploitation of creative professionals, hoping that
by the time anyone tries to stop them, it'll be too late to protect us American, or “median
humans” as a prominent AI executive likes to call us.
But with help of the FTC and others looking out for American rights, we hope that game will not
succeed. I think this panel is a great step in that direction. Regulatory agencies should act now to
protect artists, consumers, and other Americans from this unconscionable exploitation.
Regulatory agencies should demand full transparency from generative AI companies and opt-in-
only practices.
Lastly, regulatory agencies should strongly consider seeking algorithmic disgorgement on
products built on data acquired without consent, credit, or compensation, regardless whether that
company is transparent or not. Urgent measures like these will be needed to avoid, in my
opinion, the diminishing or outright destruction of most, if not all creative professional
livelihoods and the protections of all of our rights.
Thank you.
Madeleine Varner:
Thank you so much. Next, we'll hear from Steven Zapata, a concept artist and illustrator
speaking on behalf of the Concept Art Association.

Page 24 of 40
Steven Zapata:
Thank you so much and a huge thank you to the FTC for putting on this event and to all of the
attendees. I have to say the representation here is exemplary and I'm sure we are all so thankful
to see it.
My name is Steven Zapata. I'm a concept artist and illustrator, and I'm here speaking on behalf of
the Concept Art Association, an advocacy organization for artists working in entertainment,
particularly concept artists, those artists who generate striking visuals in the early design process
of an entertainment property.
CAA is committed to elevating and raising the profile of concept artists, their art, and their
involvement in the entertainment industries. We're here today because a future with unregulated
AI will hurt concept artists and all other sorts of artists across many fields. We need regulation,
intervention, and oversight. The massive datasets that text-to-image models are trained off of
contain the copyrighted works of artists, whether it be work done for ourselves or work done for
companies in a work-for-hire capacity.
To advertise our work, most of us put our art online, on social media and our personal websites.
This leaves it exposed to unethical scraping practices, which can result in AI models being
created that can mimic the signature styles of particular artists. The companies offering these
models often encourage users to request work in the styles of particular artists by name, and
many of these generations end up online with our names attached to these pieces that we didn't
make. These pieces have subject matters and qualities that we have no control over. And the
negative market implications of a potential client encountering a freely downloadable AI copycat
of us when searching our names online could be devastating to individual careers and our
industry as a whole.
Also, the nature of the typical work-for-hire contract has put us in a bad bind. Almost every
commercial artist today works under all-encompassing work-for-hire contracts that surrender all
rights to a company. And unfortunately, visual artists don't have strong union representation to
push back on this. As it stands, you can work hard for a company like Disney for 25 years and
they can take all your work, train a model on it, and then have a free and tireless replica of you to
use in perpetuity. That may currently be legal, but it certainly is not just. It is clearly unfair that
our work should be used without our permission or even our notice to train products that we will
have to compete directly against in our very marketplace.

Page 25 of 40
Given these industry pressures, artists may be coerced by clients to utilize these systems
themselves to stay up to speed with the market, thus normalizing the exploitative practices and
foundations of these models. This will inevitably damage the perception of our field and art in
general, as it will contribute to the idea that beautiful art is made easily. I can assure you, it is
not. And insofar as it is made easily by an AI, it is because it has been trained off of the beautiful
work of thousands upon thousands of artists who had to invest time and effort into creating their
art.
So we need regulation, intervention, and oversight. We as creators should have complete control
over how our work is used, but we need help. Some of the potential actions and remedies that we
hope to see include, first and foremost, ensuring that all commercial AI models utilize only
public domain content or legally licensed datasets acquired in an opt-in capacity. Opt-out is
completely insufficient here. This could mean current companies shifting to the public domain
and possibly destroying their current models in the process so that opt-in becomes the standard.
We also need transparency on datasets, and divulging your dataset should be compulsory.
Mainstream models like DALL-E 3 don't reveal their training data and don't let you search it, but
they do offer an inefficient one by one opt-out system that you can use if you think maybe your
art is in there. But because these AI systems can't unlearn, this will only remove the images from
future training datasets used by this one company and it's already too late to get out of the most
current model. Future tools that would verify compliance with future regulations will also
depend on this transparency.
We should also have AI companies pay a fine for their past practices and pay all affected artists a
fee per generation. This is to compensate artists for utilizing their works and names without
permission, should be retroactive for as long as the company has been for-profit. We must close
research to-commercial loopholes, interpreted or actual, that allow for-profit companies to
monetize the results of non-commercial research.
To close: consent, credit, compensation, control. This is what creators reasonably seek in this
new era where our work will be used to add tremendous value to these new technologies. We
need oversight. We need compulsory transparency and tools to verify compliance.
Thank you.

Page 26 of 40
Madeleine Varner:
Thank you. We're now going to hear from John K. Painting, director of the Electronic Media
Services Division of the American Federation of Musicians. John?
John K Painting:
Thank you, Maddy. Good afternoon, everyone. It is an honor to be here.
As the labor institution which represents and protects the interest of musicians who prepare and
perform instrumental music in recording studios for sound recordings, film, television and
streaming services, as well as in live theater, symphony, opera, ballet, clubs, festivals, and more
all over the US and Canada. The American Federation of Musicians is certainly no stranger to
dealing with the encroachment of technology on our profession.
But the rise of generative artificial intelligence yields a more existential fight than we have faced
before, as we approach the potential disappearance of performers livelihoods and by extension
the disappearance of a component of humanity and culture.
From our history dealing with technological advancement, like our sister entertainment unions
and guilds, the solutions sought have been traditionally approached in two ways: collective
bargaining with industry and legislative lobbying. Both paths tend to seek secondary income to
those performers whose work has been diminished by advancing technology.
For example, in 1942 with live performance and radio orchestras threatened by recorded music,
the AFM struck the record labels and eventually succeeded in establishing the Music
Performance Trust Fund, supported by label revenue that sponsors free live concerts around the
US and Canada.
More recently, the AFM has had to combat the encroachment of prerecorded music into live
engagements, such as preventing virtual orchestras from replacing pit musicians and musicals.

Page 27 of 40
We've been here before and we will be here again. Make no mistake, generative AI is coming
and it will change everything.
Back in 1927, in the face of the technological advancement of adding sound to motion pictures,
then-AFM President Joseph Weber said, "Nothing will destroy the usefulness of an organization
surer than to set its face against progress. No matter how unfavorable we may at present see the
same to our interests, we must face these changes head on rather than stick our heads in the
sand."
But if I'm a regional live theater producer, for example, with a six-figure annual budget to spend
on pit musicians, how much do you think I would be willing to pay to have a piece of technology
that replaces that cost entirely? And in that scenario, should the consumer still be paying the
same ticket prices to see that artificial musical? Is disclosure of that fact enough? What if every
theater in the country had the same technology?
Compared to previous technological shifts, the job losses here could be steep: recording and
scoring of all types can be wholesale replaced if composers and songwriters can replace the
output of live performing instrumentalists with an AI-generated facsimile. Producers may replace
the composers and songwriters themselves. Live touring across all genres would be impacted if
bands and artists could tour with AI- generated backing musicians, holographic orchestras, AI
backing tracks.
A machine-generated creation which utilizes a performer's voice, face, or in our case
instrumental sound without the actual participation of that performer in the creation is
tantamount to replacing the performer entirely. It removes the human from the creative process
and the performing process. When that happens, the work is gone, the wages are gone, the
secondary residuals are gone. But it's more than just economics. The humanity contained within
artistic and cultural expression will be gone too.
Solutions cannot merely be the bargaining of replacement remuneration if the job opportunities
are replaced wholesale, or establishing minimum hiring requirements for producers who choose
to be union signatory, or extensions of copyright law to establish a financial value to the
performers material. That may be sufficient for those whose material is being used to train the
system, but what does that mean for the young cellist who's practicing today? What will that
mean for their job prospects in five, ten, fifteen years if recording work and orchestra work dries
up?

Page 28 of 40
The solutions need to be wider than the traditional paths we've all taken owing to the cultural
damage that this problem yields. As soon as it becomes broadly accepted that art no longer
requires the involvement of human creativity, we'll have crossed the point of no return.
Consumers should not be expected to accept such an alternative at the same price points.
Musical expression transcends culture and time. It will always be a part of our society, but that
doesn't guarantee it as a viable career. The end game must be the protection of the profession.
Thanks so much for your time.
Madeleine Varner:
Thank you. And finally, we'll hear from Jen Jacobsen, executive [inaudible 01:06:37] Rights
Alliance. Jen.
Jen Jacobsen:
Thanks, Maddy. And thanks to the FTC commissioners and staff for the opportunity to
participate today. We're so grateful to the commission for your work on this issue. And I'm
honored to be here alongside all these other representatives of the creative community.
I'm Jen Jacobson, Executive Director of the Artist Rights Alliance, which is an artist-run
nonprofit that fights for the right of musicians, songwriters, and performers to be treated and paid
fairly in the digital marketplace. Like many of the groups represented here, ARA is a proud
member of the Human Artistry Campaign, a coalition of over 150 groups in 30 countries
spanning creative professions and advocating for responsible and ethical AI.
Musicians are not against AI. We don't want AI to be banned. In fact, musicians have been using
AI-driven tools for years to auto tune vocals, generate beats, assist with studio production, and in
other parts of the creative process as well. What's new today, though, are the expansive AI
models that ingest massive amounts of musical works and mimic artists voices without obtaining
creators consent or compensating them.

Page 29 of 40
Musicians want to be treated fairly, whether that means enforcing existing laws designed to
protect creators and consumers, or in some cases enacting new laws where there are specific gaps
to be filled.
Unfortunately, in today's reckless, careless rush to launch new generative AI products, we are
seeing what is euphemistically referred to as AI training or learning, but which is in fact illegal
copying of artistic works on a massive scale without consent or compensation and often without
the artist even knowing.
Of course, this is a clear infringement of creators copyrights, but it is also an “unfair and
deceptive act” that impacts both artists and consumers – and leads to unfair competition in the
music marketplace.
Musicians’ work is being stolen from them and then used to create AI-generated tracks that
directly compete with them. For example, we might see dominant streaming platforms packing
playlists with AI music that they obtain free of charge or at a massive discount, which then
lowers their own royalty obligations and diminishes artists wages.
The increasing scale of machine-generated music dilutes the market and makes it more difficult
for consumers to find the artists they want to hear. It makes it harder for artists to connect with
their fans, and it devalues human creativity.
And perhaps even more disturbingly, AI models are now using artists’ faces, voices, and
performances without permission to make digital impersonations that not only create consumer
confusion, but also cause serious harm to both fans and artists. These deep fakes have depicted a
band canceling a concert that wasn't actually canceled. They've shown artists selling products
that the artists never endorsed. We've seen false depictions of musicians badmouthing their own
fans. This isn't a hypothetical harm. This type of consumer deception and fraud are happening
right now. It's hard to imagine anything more personal to an artist or to anyone than being
depicted as doing or saying things that they would never do or say. It's not only confusing to
fans, but humiliating to the artists themselves and undermines their public image.
This conduct may violate “right of publicity” laws in several states, but it needs to be recognized
universally as a misappropriation that causes real harm not only to the artists, but to the entire
market by confusing consumers and creating unfair competition. And no one is as well-
positioned to protect consumers and fans in all fifty states than this agency.

Page 30 of 40
The final point I want to make is about the importance of choice. Many AI companies who have
illegally vacuumed up hundreds of thousands of musical works and recordings now say that
artists can simply contact the company and “opt out.” This is essentially proposing a new form of
uncompensated labor that musicians and composers have to perform if they want to avoid
exploitation. It's also completely impractical given the proliferation of new services ingesting
work without permission. Such burden shifting is not only unfair, it is morally wrong and
antithetical to basic principles of artistic integrity. Artists have the right to control whether and
when their work is being used, especially when these uses compete with their own livelihood or
violate their own values.
Art is about a human-to-human connection, about sharing emotions and lived experiences.
Machines can't share emotions or lived experiences because they haven't had any. Only humans
can do that. For musicians, like all the other creators represented here today, the unethical use of
AI poses an existential threat to our livelihood. But for all of us, the very foundations of human
creativity and culture are at stake. It's hard to imagine anything more important than that.
Thank you again for including ARA in this discussion, and I look forward to answering any
questions.
Madeleine Varner:
Thank you, Jen. And thank you everybody for your remarks. So now we're going to move into a
Q and A portion. And I just wanted to note too that Duncan Crabtree-Ireland of SAG-AFTRA
had to leave a little early to return back to the bargaining table.
So to get started, I'm going to open with the first question.
Recently there's been some attention paid to the idea of offering opt-outs to artists who don't
want to particular company from using their works to train generative AI models. Do such
measures offer sufficient protections for creators? I'm going to go first to John August from the
Writers Guild of America West. John?
John August:

Page 31 of 40
Thank you. As Chair Khan said in our opening remarks, the advent of AI doesn't change
fundamental ways that the world is supposed to work, and that world works in permission first.
That's why we think it's critical that we require artists have affirmative consent before the work
can be used to train generative AI models and that they have to be compensated fairly when they
do so. The same should be true for all artists, including artists like us who do work for hire and
don't hold the copyright on our work.
And this system needs to be opt-in and not opt-out. As Jen just said, there are so many
companies out there developing and training AI models, to be forced to continually track all of
them down to opt out is an enormous administrative burden on individual artists. It's not
practical. It has to be opt-in rather than opt-out.
Madeleine Varner:
Thank you. Karla, I'm curious about your thoughts on this.
Karla Ortiz:
Yeah, so opt-out is an ineffective and inappropriate standard for commercial use of copyrighted
works including a generative AI. Once a model is trained on data, it cannot be deleted unless the
whole model is retrained from scratch. By the time a model is made public, it's already too late to
opt out. Number two, most AI companies keep that training data secret, preventing artists from
even knowing if their works were used to train a model. Number three, existing opt-out
procedures often ask users to list works used to train the model they own, but as we just
mentioned, that training data is secret, so it's an impossible task. And four, there are hundreds of
AI models already in the market and more.
Does that mean we have to opt out on each and every one of them? That's a full-time job. What
about if those models update? What about if they don't publicize and they use third parties? What
if those models in the opt-out forms are not an artist's native language? What about artists who
never spend time online or don't even know this is happening?
Basically, tech companies must respect artists ownership rights. These seeking to profit from
others works should have the burden of obtaining permission. Explicit opt-in is the only way
forward. It's really how we ensure generative AI models exclude unauthorized works from the
beginning. Thank you.

Page 32 of 40
Madeleine Varner:
Thank you. Neil Clarke of Clarkesworld, I'd love to hear your thoughts on this as well.
Neil Clarke:
I wholly agree with all the things that have been thrown on the table there. This is what you get
when the fox designs the chicken coop. The wrong people are making the decisions when you
have something like this happening. And there's no standard, there's no consequence for them
ignoring these things at the moment either. So we really need to have them seeking consent,
explicit consent, to include these works.
Madeleine Varner:
Thank you. I'm going to move on to our next question. Like “opt-outs,” the idea of licensing out
work for training purposes either by individual creators or others who may retain rights to said
work has garnered attention. What interests or concerns do you have with this kind of approach?
And I'm going to kick it to Jen Jacobson of ARA first. Jen?
Jen Jacobsen:
Thanks Maddy. ARA does believe that it's possible to have a market for licensing our work to
responsible, ethical AI developers. The entire music streaming market is based on licensing, so
we know it can work. But there are three things I would say about it.
First, artists need to have complete opt-in choice, as we've just heard, about whether to license
their work or not.
Second is artists need to have enough power to negotiate fair license terms with these gigantic AI
developers. There's actually a bill in Congress sponsored by representative Deborah Ross that
would give small and independent musicians an antitrust exemption so they can come together
and negotiate collectively, both with AI developers and streaming platforms, which is something
they do not have the leverage to do currently.

Page 33 of 40
And the third thing I would say is there is not a one-size-fits-all licensing system that will work
for all creators or even for all musicians. Assuming there is a level playing field for negotiating,
we think the best way for musicians to license their work is in the free market, which may look
different for every use, every artist and every company.
Madeleine Varner:
Thank you. Steven Zapata. I'd love to hear your thoughts on this.
Steven Zapata:
Without a doubt, licensing will be essential in the future, but we must accomplish that through an
opt-in system, otherwise there would be no real negotiating leverage for creators. And the focus
of licensing, I think should go towards new opt-in foundation models, not the fine tuning of
existing unethical models. As to when companies hold the rights to work done for hire and want
to license or train off of that, we need regulation.
I mean, visual artists, for example, lack the union representation to push back against contracts
that claim all ownership. And without regulation, I think predatory contracts will just run
rampant in this sector. And collective licensing is also troubling. The early experiments we are
seeing with "contributor funds" from companies like Shutterstock are paying out less than
pennies. I mean actual fractions of a penny per used image. That's all they want to pay a creator
for their life's work and to create a tool that will directly compete against them forever.
If that's not unfair, I don't know what is.
Madeleine Varner:
Thank you. John K. Painting of AFM. I'm curious about your thoughts.
John K Painting:
Sure. This goes back a bit to part of the solution I mentioned in my remarks. Let's say I'm the
drummer for Taylor Swift and her catalog is used to train a generative AI system that can pump
out an album on a prompt, like give me an album of Christmas songs in the style of Taylor Swift,

Page 34 of 40
then the drums on that generated album must have used my sound to mimic the style. So not only
should I have had the option to opt in there, of course, but I should see some form of benefit or
compensation for that because those new parts are clearly copying mine.
But this is still only a bandage on the problem, because if this scenario works really well, it likely
means that I'm not getting hired to record any new albums anymore because this system can just
pump this stuff out. So do these new albums have the same value? Should they cost the same to
buy and stream? Does the art have the same meaning and the same impact? That's all the
existential crisis that we're facing right now.
Madeleine Varner:
Thank you. So I'm going to move on to our next question, which is: What kind of insight do you
feel like you have now into how your work or likeness is being used by generative AI systems,
and what kind of transparency do you feel is needed?
And Umair from Authors Guild, I'm going to start with you.
Umair Kazi:
Thanks, Maddy. Our members are keenly aware that their works are being used by AI systems.
We get reports. Our members try out different prompts and AI systems and chat box reveal
details about their works. And of course, several of our members have also reported finding their
books in Books3, which is a data set of containing 200,000 books that's downloaded from a
pirate source, which was used to train Meta's LLaMA, Bloomberg's GPT, and others.
But Books3 is kind of an anomaly, is that it was publicly downloadable and its contents were
visible and searchable. There is a lack of transparency from AI developers about training data
sets, which makes it very difficult to ascertain which works were actually used to train the
models and how.
Much of the information about ingestion comes from the prompting experiments that I
mentioned. We and other authors have been able to prompt, for instance, GPT to produce
extensive detailed summaries of works and text in the style of and even incredibly compelling
outlines for possible derivative works like sequels using settings, characters.

Page 35 of 40
As far as what we want, we want AI companies to be required to fully disclose the complete lists
of copyrighted works, books in particular is the medium that we deal with most often, that are in
the training data sets or provide specific links to where the data sets were obtained from. And
anyone compiling a training dataset should be similarly obligated to disclose the sources. And in
the case of copyrighted works, a complete list of works that have been included in the dataset.
We also think authors and artists should have the right to say that they don't want their identities,
works, voice or style used in outputs. And we believe that there should be a requirement to
conspicuously label fully or substantially AI generated words in online marketplaces.
Madeleine Varner:
Doug Preston, I'm curious to hear your thoughts on this question.
Douglas Preston:
Yes, thank you. As a plaintiff in this lawsuit, one of the things that I did was to go to ChatGPT-3
and ask it many detailed questions about my books, my characters, very minor characters that
have never appeared in a book review or in Wikipedia, and it was absolutely extraordinary how
knowledgeable ChatGPT was. It was able to write a whole paragraphs in my style, using my
characters, using my settings, using everything in my books, and yet it was all stolen from me
and fake and AI generated. And that was very disturbing. And I also used that Books3 database,
that dataset that Umair was speaking about, and found that that dataset has all my books, all 40
of them on it, including many different editions. That was another shock.
But as far as how we know our books are being used in AI training, we have absolutely no idea.
It's a black hole. OpenAI is training ChatGPT 5 right now, is building it. They refuse to answer
any questions from the Author's Guild about what data sets they're using, where they're getting
their books, and how they're being used. There's no transparency at all. It's an absolute black
hole. All we want is control of our creative work, fair compensation and opt-in consent.
Madeleine Varner:
Sara Ziff, I'm curious about your thoughts on this question.

Page 36 of 40
Sara Ziff:
Yeah, thank you. Models have very little insight into how their work or likeness is being used in
general, let alone in the context of generative AI. Normally they don't see their contracts with the
brands and often don't know how their image will be used, whether how much they'll be paid. So
generative AI introduces the potential for further exploitation in an already exploitative work
environment. In terms of the transparency we want, I think some key elements would include
requirements for explicit consent, notification of use, compensation, and liability for
misrepresentation.
Madeleine Varner:
Thank you. Bradley Kuhn, I'm curious about your thoughts on this question.
Bradley Kuhn:
Well, first of all, there's now really no question that the body of Copyleft in FOSS is a huge part
of the software assisted development machine learning systems such as Microsoft's GitHub
CoPilot. We've exposed many of these examples, but the companies are also playing cat and
mouse by simply excluding, on the generative backend, the most egregious examples of
copyright infringement when they're publicly called out.
In my opinion, there's no reason that big tech shouldn't be regulated to make these systems
transparent, completely end to end. In my view, the public should have access to the input set,
have access to the source code of the software that does the training and generation, and most
importantly, access to the source code that does these forms of backend generation exclusion, the
latter of which I think would expose the duplicity of big tech's policies here.
Finally, I expect that once we have real transparency, it will bear out what many of the other
speakers have noted today, that the issues of machine learning systems can't be solved merely
with a financial compensation model to creators. FOSS really shows this explicitly since most
FOSS authors are doing the writing of the software altruistically and the compensation that
they're seeking is the requirement for future improvements to the comments, not a financial
compensation. So we really need full transparency in these systems to assure that essential non-
monetary policy, licensing term and consumer's rights are upheld.
Madeleine Varner:

Page 37 of 40
Thank you. So I'm going to ask our final question for the day, which is, you've all described the
harm to creators and their professional interests when companies use generative AI tools in
certain ways, such as when those tools rely on the use of protected works or people's voices and
images. Can those uses also wind up harming consumers? And if so, how?
And for this question, we're going to go to Tim Friedlander from the National Association of
Voice Actors. Tim?
Tim Friedlander:
Cool. Thank you for this question. The ability to create a synthetic voice from anyone who has
recorded audio is easy and simple and dangerous. Currently now it only takes three seconds of
source audio to create a realistic voice clone. And this synthetic content can be used to deceive
consumers into believing that a trusted voice is communicating with them. This can lead to
relying on false and misleading information and potentially even implicate the human whose
voice has been used to harm people. In the last couple of days, we've seen Tom Hanks and
MrBeast release statements saying that there is synthetic versions of their image and voice cell in
ads they did not authorize. And currently there's a show streaming on a major media streaming
channel that is using AI generated voice dubs for their English dubs.
At a minimum the consumers should know when AI is used to generate voices and receive a
warning that the information they're going to receive may not be accurate. We're also coming
into an election, and we know that synthetic voices will be used, they have been used, and it's
possible for anybody to use a synthetic voice for disinformation. And no one, a voice actor or
anybody in general, wants to be the recognizable voice of disinformation in this coming election.
I want to leave just with a personal thought that we all know the sound of our friends and family
and our loved ones' voices. You call somebody and say, "Hi, it's me." We know who we're
speaking to. I spoke with a friend of mine, his name's Patrick, last night. Two months ago or a
few months ago, he received a phone call from somebody claiming to have kidnapped his
daughter, was demanding ransom, played a recording or a voice of his daughter screaming,
"Papa, help me. Help me. Papa, save me." He said, "I knew deep in my bones, that was my
daughter's voice." His daughter was safe. It was a scam, it was a fake. But he said "The voice
that I heard on that line was my daughter's voice. It was the same voice I heard when I dropped
her off at school. And it was the same voice that I heard when I picked her up when I found out
she was safe." The danger to consumers is real. It is tangible. And it is here now.
Madeleine Varner:

Page 38 of 40
Thank you. And that concludes our Q&A portion. Now we're going to turn it over to
Commissioner Bedoya for some closing remarks. Over to you, Commissioner.
Commissioner Alvaro Bedoya:
Hi everyone. Can you hear me and see me? Wonderful. Tim, what a note to end on. As a parent,
that is chilling. And I just want to say before I begin, I want to touch on two issues that I think
hopefully will compliment some of the conversations that have been had today. I'm profoundly
worried about fraud and AI use. I'm profoundly worried about what's happening to your work,
your property, the fruit of your minds, quite literally.
I just want to offer two thoughts.
The first doesn't really have to do with the law. It has to do more with first principles and really
what's at stake. And my hope is that this audience in particular, might think that important. The
second does have to do with a very specific legal application of generative AI that concerns me
profoundly and that I recently wrote about in the LA Times on Labor Day. But let's first talk
about what's at stake here.
There is this foundational moment in American letters. I think it was 1837, Ralph Waldo
Emerson gives this talk called the American Scholar. He goes on about how for way too long
American writers, American poets have looked to Europe. He basically calls out writing for
being boring. He says, he has this wonderful line, "Genius is always the enemy of genius by over
influence." And he says that for 200 years, English poets have been Shakespeare-ized, basically
saying that people have been copying the work of this genius rather than creating new work. And
he said, "We need an American voice. We need an American scholar. It needs to be rooted in the
soil, it needs to be free of these European strictures." And sitting in the audience that day was a
journeyman printer from Brooklyn named Walter Whitman.
One of the great mysteries of American letters is that basically out of nowhere in 1855, a
basically middling journeyman printer and occasional journalist creates what is arguably the
greatest work of American letters, “Leaves the Grass.” This collection turned poetry itself on its
head. Previously, poetry had restricted itself to certain strictures around rhyme and meter.
Previously, yes, poetry maybe touched upon love and human sexuality, but it was always demure
to a certain degree, think Keats' “Ode to a Grecian Urn.” And suddenly everything was possible.
Suddenly Whitman made it acceptable for people to write in free verse. Whitman made it
acceptable to write about love and sexuality in a way that was previously considered completely
not on the table.

Page 39 of 40
There is no Allen Ginsburg without Whitman. There is no Gabrielle Garcia Lorca without
Whitman.
When I think about the work all of you do and I think about the challenges we're confronting,
and you think, what can we expect from generative AI? I think there are people who basically
expect everything, expect the world, expect there are no limits whatsoever on what is possible.
And what I think it's important to remember is that because of the fundamentally recursive nature
of what we call artificial intelligence, where it takes what it's read, what it's seen and returns it to
you, I think that it is wildly unrealistic and frankly dangerous to think that it will break every
bound of human creativity.
I think that, to paraphrase Emerson, what can we expect? We can expect a lot of bad copies of
Shakespeare, but good luck trying to create the next Walt Whitman. I don't think it will ever
extinguish the genius of human creativity. And it's important we disabuse people of the notion
that it remotely will. And so, as a question of first principles, I think we need to remember that
the only place genius comes from is people, is humans. And any effort to persuade the world
otherwise will fail. But we need to make sure to remind people of that.
Let me get off my rant for a moment and speak about one specific legal application, which is
something that concerns me, which is the capture of the art of actors and the art of writers. A
couple weeks ago I started digging into this and I saw two things.
First of all, I was reading all these industry analyses about what AI promised for industry, and
they were all over the place, but the one thing they have in common is they say, to the degree
that work requires social and emotional skills, artificial intelligence is going to struggle.
I was reading these reviews from very fancy think tanks about that. And then Justine Bateman
posted a note on social media. She said, "Hey, background actors, I hear some of you have been
scanned on the lot. Why don't you share your stories with the world?" And these stories started
coming in. "Yeah, hey, I was on the lot. And on day three with no notice, I was told to go to that
tent over there. I walked into that tent and I was scanned from 360 degrees by 80 cameras. They
paid me $25. They told me that if I said no, I'd get fired." Someone else writing about, same
story, surprise request to get scanned, surprise capturing not just in many instances of people's
bodies, but also their emotions.

Page 40 of 40
Hearing these stories reminded me in Congress perhaps of 1914, when Congress passed the law
that created this commission that all of you are speaking before today, the Federal Trade
Commission. In 1914, Congress had the choice of passing a law that specifically enumerated acts
1, 2, 3, 4, 5, 6, 7 that would be illegal. One House of Congress actually considered that, actually
voted on it. Then they stopped and said, no, we need to consider the fact that there will be
innovation in unfair methods of competition. We need to create an institution that is free to meet
the innovations of large, powerful entities that will stifle competition in American industry in
whatever corner it may be found.
And when I hear about writers who worry, new writers, young writers worry that the moment I
arrive, I'm going to be asked to feed my scripts in, to train a new AI. When I hear about
background actors, young actors, how lots of future actors are discovered, but who are the least
powerful, least experienced, least savvy of all actors being forced to get scanned in the nude
sometimes or in other really uncomfortable situations, it strikes me as more than innovative and
it fills me with concern.
I will say, because this is a law enforcement agency, these are allegations. I did not investigate
these things, but the shape of what I'm seeing concerns me profoundly. And you have to know
that in myself, in our chair, in my colleague, Commissioner Slaughter, you have public servants
who are keenly interested in your stories and what's happening to you and what's happening to
your work. So with that, I will end and I'm not sure, actually, I did read the script. I don't know if
I wrap everything or I pass it back to you, Madeline. Let me pass it back to you.
Madeleine Varner:
Thank you so much, Commissioner Bedoya. On behalf of the FTC, I'd like to thank everybody
for a robust discussion, for sharing your perspectives and experiences and for raising more
awareness on these issues. We're sincerely so grateful.
We will be posting the video and transcript of this round table on the event webpage. And in the
meantime, thank you again so much to all the participants as well as the folks who are watching
right now.
We hope you all have a great rest of your day. Thank you.