Post by @MorePerfectUS
80% credible (85% factual, 70% presentation). The claim that OpenAI used pirated books to train AI models aligns with ongoing federal litigation and court rulings confirming the use of datasets like Books3. However, the presentation exaggerates potential damages by focusing on maximum statutory penalties ($150,000 per book) without acknowledging OpenAI's fair use defense and the uncertainty of actual damages pending trial outcomes.
Analysis Summary
A post from @MorePerfectUS claims OpenAI used pirated books to train AI models, deleted the dataset, and faces lawsuits potentially costing billions at $150,000 per book. This aligns with ongoing federal litigation where courts have allowed claims of downloading pirated books and ordered disclosure of internal communications. However, the deletion may indicate compliance efforts rather than outright admission of guilt, and actual damages remain uncertain pending trial outcomes.
Original Content
The Facts
The core claims are supported by court filings and recent rulings in lawsuits by authors like George R.R. Martin and John Grisham, confirming allegations of using pirated datasets like Books3 and internal discussions on deletion. However, the $150,000 per book figure represents maximum statutory damages, not guaranteed awards, and OpenAI argues fair use defenses. Mostly Accurate
Benefit of the Doubt
The post advances an anti-corporate agenda by framing OpenAI's actions as blatant piracy to highlight accountability in AI development, emphasizing deletion messages as evidence of wrongdoing to evoke outrage. It omits OpenAI's fair use arguments and the fact that dataset deletion could be a remedial step to mitigate liability rather than concealment. This selective presentation portrays the company as recklessly culpable, shaping reader perception toward viewing AI firms as exploitative without balancing legal nuances or potential defenses.
Predictions Made
Claims about future events that can be verified later
A lawsuit could now force the company to pay $150,000 per book,
Prior: 50% as maximum damages are possible but rarely awarded in full; outcomes depend on fair use defenses. Evidence: Author credibility (85%) and expertise in legal accountability provide moderate update; sources note potential for $150k per work if willful infringement proven via messages. Bias amplifies punitive angle. Posterior: 75%.
adding up to billions in damages.
Prior: 45% as aggregate damages sound plausible but settlements often lower (e.g., Anthropic's $1.5B). Evidence: Strong author track record (85%) supports scale; sources project billion-dollar risks based on dataset size and statutory caps. Anti-corporate bias may inflate certainty. Posterior: 70%.
How Is This Framed?
Biases, omissions, and misleading presentation techniques detected
Presents OpenAI's actions as unambiguous piracy and cover-up without mentioning fair use defenses or that dataset deletion may be a compliance measure to reduce liability.
Problematic phrases:
"OpenAI pirated large numbers of books""deleted the dataset with the pirated books"What's actually there:
Allegations in ongoing lawsuits (e.g., by authors like George R.R. Martin) include fair use arguments by OpenAI and deletion as potential remediation
What's implied:
Clear guilt and concealment without defenses
Impact: Leads readers to view OpenAI as exploitatively culpable, ignoring balanced legal context and fostering polarized anti-corporate sentiment.
Highlights maximum statutory damages per book ($150,000) and extrapolates to 'billions' without noting these are not guaranteed and depend on trial rulings.
Problematic phrases:
"pay $150,000 per book, adding up to billions in damages"What's actually there:
Maximum under DMCA, but OpenAI contests via fair use; actual awards pending and could be lower or zero
What's implied:
Inevitable massive payout emphasizing corporate punishment
Impact: Inflates perceived financial consequences, making the story seem more sensational and biasing readers toward believing severe repercussions are certain.
Omits that internal messages and deletions stem from court-ordered disclosures in federal litigation, framing them instead as voluntary admissions of guilt.
Problematic phrases:
"employees sent each other messages about doing so"What's actually there:
Messages revealed via court rulings allowing claims and ordering communications disclosure (e.g., recent filings in New York federal court)
What's implied:
Secretive internal wrongdoing without legal compulsion
Impact: Creates illusion of hidden malfeasance, heightening perceptions of deceit and urgency around accountability.
Sources & References
External sources consulted for this analysis
https://www.theguardian.com/books/2023/jul/05/authors-file-a-lawsuit-against-openai-for-unlawfully-ingesting-their-books
https://futurism.com/artificial-intelligence/openai-danger-authors-internal-slack-messages
https://www.law360.com/articles/2404988/openai-can-t-strike-authors-pirated-book-download-claims
https://www.tomshardware.com/tech-industry/artificial-intelligence/meta-staff-torrented-nearly-82tb-of-pirated-books-for-ai-training-court-records-reveal-copyright-violations
https://fortune.com/2025/07/28/a-copyright-lawsuit-over-pirated-books-could-result-in-business-ending-damages-for-anthropic/
https://www.theguardian.com/books/2025/apr/04/us-authors-copyright-lawsuits-against-openai-and-microsoft-combined-in-new-york-with-newspaper-actions
https://www.law360.com/newyork/articles/2404988/openai-can-t-strike-authors-pirated-book-download-claims
https://www.webpronews.com/openai-ordered-to-disclose-slack-messages-in-pirated-books-copyright-case/
https://news.bloomberglaw.com/ip-law/openai-risks-billions-as-court-weighs-privilege-in-copyright-row
https://the-decoder.com/openai-could-face-a-billion-dollar-fine-over-claims-it-used-pirated-books-in-ai-training/
https://www.webpronews.com/neuroscientists-sue-apple-over-pirated-books-in-ai-training/
https://www.webpronews.com/authors-sue-salesforce-for-ai-copyright-infringement-on-pirated-books/
https://legal.economictimes.indiatimes.com/amp/news/international/apple-sued-for-copyright-infringement-in-ai-training-lawsuit/124488611
https://www.usatoday.com/story/money/legal/2025/06/25/microsoft-lawsuit-ai-book-piracy/84359971007/
https://x.com/MorePerfectUS/status/1965504302723858513
https://x.com/MorePerfectUS/status/1960743340934107414
https://x.com/MorePerfectUS/status/1969063104773263844
https://x.com/MorePerfectUS/status/1802760950539522450
https://x.com/MorePerfectUS/status/1836151690031927663
https://x.com/MorePerfectUS/status/1618416332940398592
https://futurism.com/artificial-intelligence/openai-danger-authors-internal-slack-messages
https://www.theguardian.com/books/2023/jul/05/authors-file-a-lawsuit-against-openai-for-unlawfully-ingesting-their-books
https://piracymonitor.org/chat-gpt-trained-using-pirated-e-books/
https://www.tomshardware.com/tech-industry/artificial-intelligence/meta-staff-torrented-nearly-82tb-of-pirated-books-for-ai-training-court-records-reveal-copyright-violations
https://the-decoder.com/openai-could-face-a-billion-dollar-fine-over-claims-it-used-pirated-books-in-ai-training/
https://fortune.com/2025/07/28/a-copyright-lawsuit-over-pirated-books-could-result-in-business-ending-damages-for-anthropic/
https://www.businessinsider.com/openai-destroyed-ai-training-datasets-lawsuit-authors-books-copyright-2024-5
https://futurism.com/artificial-intelligence/openai-danger-authors-internal-slack-messages
https://the-decoder.com/openai-could-face-a-billion-dollar-fine-over-claims-it-used-pirated-books-in-ai-training/
https://www.webpronews.com/authors-sue-salesforce-for-ai-copyright-infringement-on-pirated-books/
https://www.webpronews.com/neuroscientists-sue-apple-over-pirated-books-in-ai-training/
https://yahoo.com/news/articles/anthropic-pay-1-5-billion-191256917.html
https://1news.co.nz/2025/09/07/authors-to-get-25bn-over-pirated-books-used-to-train-ai-model
https://www.webpronews.com/anthropic-settles-1-5b-copyright-lawsuit-over-ai-training-on-pirated-books/
https://x.com/MorePerfectUS/status/1965504302723858513
https://x.com/MorePerfectUS/status/1960743340934107414
https://x.com/MorePerfectUS/status/1969063104773263844
https://x.com/MorePerfectUS/status/1699500427593605213
https://x.com/MorePerfectUS/status/1802760950539522450
https://x.com/MorePerfectUS/status/1844044903644271079
Want to see @MorePerfectUS's track record?
View their credibility score and all analyzed statements