Tech Platforms - December 5, 2025
Who's suing AI and who's signing: The New York Times and Chicago Tribune just sued Perplexity. Meta inked fresh content deals with CNN, Fox News, People Inc, Le Monde, USA Today and more. Getty failed to secure a UK copyright precedent against Stability AI. And Penske Media became the first publisher to sue Google over AI Overviews' impact on traffic and revenue.
If you build, ship, market, or protect products, this isn't background noise. It's the rulebook forming in real time-defining what "training," "RAG," attribution, and compensation will look like across AI products.
Snapshot: the latest moves
- NYT vs Perplexity (Dec 5): The Times says Perplexity uses retrieval-augmented generation to copy and deliver its journalism, including content behind its paywall, without permission or payment.
- Chicago Tribune vs Perplexity (Dec 5): Claims unlawful scraping and near-verbatim outputs; adds trademark claims tied to hallucinations and misleading omissions.
- Penske Media vs Google (Sep 15): First publisher suit over AI Overviews, alleging reduced clicks and affiliate revenue, and no way to opt out of AI use while staying in search.
- US News & World Report vs OpenAI (Nov 26): Alleges training on and regurgitation of its content without credit or compensation; also points to misattributed inaccuracies.
- MediaNews Group regionals vs OpenAI/Microsoft (Nov 26): Nine Alden papers seek $10bn+ in damages over alleged mass copying; eight Alden titles already sued in Apr 2024.
- Getty Images vs Stability AI (UK update, Nov 4): Training claim withdrawn; Stability largely vindicated by High Court. No sweeping UK precedent on training. US case continues.
Also in the courts
- Encyclopedia Britannica & Merriam-Webster vs Perplexity (Sep 12): Claims "free riding" via summaries that divert traffic.
- Folha vs OpenAI (Aug 29): Argues unauthorized use, asks to stop use and destroy models trained on its content.
- Yomiuri Shimbun vs Perplexity (Aug 12): Seeks damages for alleged use of 100k+ articles and images; flags lost ad revenue from fewer clickthroughs.
- Reddit vs Anthropic (Jun 4): Claims 100k+ unauthorized accesses despite bot blocks.
- Ziff Davis vs OpenAI (Apr 25): Says models were trained on its properties (IGN, PCMag, etc.) without a license.
- News/Media Alliance membership vs Cohere (Feb 13): First round win for publishers; suit alleges paywall bypass, verbatim regurgitation, and harmful hallucinations.
- Indian publishers vs OpenAI (Jan 28): Group action claims scraping and adaptation of content; jurisdiction contested.
- Canadian coalition vs OpenAI (Nov 29, 2024): Major publishers allege systematic infringement.
- News Corp (WSJ, NY Post) vs Perplexity (Oct 21, 2024): Copyright and trademark suit alleging "massive freeriding."
- Mumsnet vs OpenAI (Jul 19, 2024): Legal action over scraping six billion+ words and terms breaches.
- CIR (Mother Jones, Reveal) vs OpenAI/Microsoft (Jun 28, 2024): Claims use without permission or compensation.
- The Intercept, Raw Story, AlterNet vs OpenAI (+ Microsoft for The Intercept) (Feb 28, 2024)
- NYT vs OpenAI/Microsoft (Dec 27, 2023): Seeks destruction of models trained on Times content; OpenAI disputes usage and methods of eliciting verbatim text.
Licensing: where content is being paid for
- Meta: New deals with CNN, Fox News, People Inc, Le Monde, USA Today Network, The Daily Caller, The Washington Examiner; earlier agreements include Reuters.
- Microsoft: People Inc joins the upcoming Publisher Content Marketplace; FT, Reuters, Axel Springer, Hearst Mags, USA Today Network, Informa also listed partners.
- OpenAI: The Washington Post, Guardian, Schibsted, Axios, Future, The Independent, LA Times, Dotdash Meredith (People Inc), Vox Media, The Atlantic, News Corp, Time, Financial Times, Le Monde, Prisa Media, Associated Press, Conde Nast.
- Amazon: AI deals with The New York Times, Conde Nast, Hearst (Rufus integration and training use).
- Perplexity: Gannett/USA Today Network; The Independent, LA Times, Lee Enterprises; Time, Der Spiegel, Fortune, Texas Tribune and more; plus an image display deal with Getty.
- Prorata.ai (Gist.ai): 500+ publications including The Atlantic, Time, Fortune, Fast Company, Vox, The Verge, Guardian, Daily Mail, Sky News; revenue sharing via attribution tech; NMA partnership in place.
- Google: Associated Press real-time feed into Gemini.
- Mistral: AFP multi-year, multi-language deal for Le Chat.
- Synthesia: Research license with Shutterstock for video/avatar model pre-training.
Money on the table
Reports suggest OpenAI has been offering $1m-$5m per year to license news content, with one News Corp deal reportedly worth $250m over five years. Amazon is reportedly paying The New York Times $20m-$25m per year.
Some publishers refuse to sign for now. Others lean in-seeking attribution, links, and model-training fees. Most deals mix real-time display rights, API/enterprise seats, and limited training rights with attribution requirements.
Why this matters (by team)
Legal
- Clarify definitions in contracts: "training," "fine-tuning," "RAG," "indexing," and "caching." Disallow paywall bypass and require deletion on breach.
- Demand indemnities for hallucinations, misuse of marks, and right to audit logs. Set clear remedies and per-incident penalties.
- Maintain takedown workflows under DMCA/notice regimes; track bot access vs. robots/terms.
IT & Development
- Harden access controls on paywalled content. Enforce robots.txt, rate limiting, and WAF rules; monitor scraper fingerprints and ASN patterns.
- Add "noai" directives and structured attribution hints; log RAG-like fetches hitting your endpoints.
- Instrument link prominence tracking for AI referrals to validate deal value.
Product
- Quantify tradeoffs: model training rights vs. real-time answer syndication vs. traffic cannibalization.
- Negotiate link placement, logo usage, and analytics access. Plan A/B tests across partners.
- Build AI-native features: summaries, audio, Q&A, and enterprise tools that deepen retention.
Marketing, PR & Comms
- Demand visible attribution and top-of-answer links. Track changes in branded search and affiliate revenue.
- Prep responses for hallucinations misattributed to your brand; clarify approval pathways with partners.
Editorial & Writers
- Align on what gets syndicated and what stays subscriber-exclusive. Require citation and recency labels.
- Surface pillar content and evergreen updates to maximize high-intent referrals from AI assistants.
Playbook: actions to take now
- Run a crawler audit. Map bots vs. terms vs. robots.txt; tighten WAF and rate limits where needed.
- Segment rights. Separate display rights from training rights; price them differently; require deletion on contract end.
- Negotiate link prominence. "Above the fold," logo use, and multiple inline links, not footnotes.
- Set hallucination guardrails. Require rapid correction paths, visible disclaimers, and per-incident remedies.
- Track ROI. Build dashboards for AI referrals, CTR, dwell, conversions, and affiliate deltas.
- Protect marks. Explicit rules on brand usage, style, and no synthetic endorsements.
- Create a standard license addendum. Terms, tech specs, audit rights, and SLA for takedowns.
Quick reference: publishers suing AI platforms
- New York Times vs Perplexity
- Chicago Tribune vs Perplexity
- US News & World Report vs OpenAI
- Nine MediaNews Group (Alden) papers vs OpenAI/Microsoft
- Penske Media vs Google
- Encyclopedia Britannica & Merriam-Webster vs Perplexity
- Folha vs OpenAI
- Yomiuri Shimbun vs Perplexity
- Reddit vs Anthropic
- Ziff Davis vs OpenAI
- News/Media Alliance members vs Cohere
- Indian publishers vs OpenAI
- Canadian coalition vs OpenAI
- News Corp vs Perplexity
- Mumsnet vs OpenAI
- The Center for Investigative Reporting vs OpenAI/Microsoft
- Eight Alden daily newspapers vs OpenAI/Microsoft
- The Intercept, Raw Story, AlterNet vs OpenAI (+ Microsoft for The Intercept)
- The New York Times vs OpenAI/Microsoft
- Getty Images vs Stability AI (UK outcome mixed; US case continues)
Quick reference: publishers signing AI deals
- CNN, Fox News, People Inc, Le Monde, USA Today Network, The Daily Caller, The Washington Examiner - Meta
- People Inc - Microsoft
- Reuters - Meta
- Getty - Perplexity (image display)
- Gannett/USA Today Network - Perplexity
- Conde Nast, Hearst - Amazon (Rufus)
- 500+ publications - Prorata.ai (Gist.ai)
- The New York Times - Amazon
- The Washington Post - OpenAI
- Shutterstock - Synthesia (research license)
- News/Media Alliance - Prorata.ai
- Guardian - OpenAI
- Schibsted - OpenAI
- AFP - Mistral
- Associated Press - Google (Gemini feed) and OpenAI (earlier)
- Axios - OpenAI
- Future - OpenAI
- The Independent, LA Times, Lee Enterprises - Perplexity
- DMG Media, Guardian, Sky News, Prospect - Prorata.ai
- Hearst - OpenAI
- FT, Reuters, Axel Springer, Hearst Mags, USA Today Network - Microsoft
- Conde Nast - OpenAI
- FT, Axel Springer, The Atlantic, Fortune - Prorata.ai
- Time, Der Spiegel, Fortune, Texas Tribune and more - Perplexity
- Time - OpenAI
- Vox Media - OpenAI
- The Atlantic - OpenAI
- News Corp - OpenAI
- Dotdash Meredith (People Inc) - OpenAI
- Informa - Microsoft
- Axel Springer - Microsoft and OpenAI
- Financial Times - OpenAI
- Le Monde, Prisa Media - OpenAI
- Shutterstock - OpenAI
Two signals worth watching
- Google AI Overviews: If courts decide summaries reduce fair referral traffic without a clear opt-out, the economics of publishing shifts again.
- UK litigation trajectory: Getty's UK outcome avoided a sweeping ruling on training. Expect more narrow, feature-specific findings unless legislators step in.
Useful references
Level up your team's AI fluency
If you're evaluating licenses, integrating AI features, or setting policy, structured training helps. Explore role-based paths and company-specific tracks below.
Get Daily AI News
Your membership also unlocks:
700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)
Advertisement
