Over the past few weeks @aisvarya17.bsky.social and I have been making some tweaks to the Deals and Disputes tracker we maintain for the @towcenter.bsky.social. The biggest update: you can now filter the deals and lawsuits by country! tow.cjr.org/ai-deals-law...

Latest updates to @towcenter.bsky.social's Deals & Disputes Tracker: 💰 OpenAI commits additional funding to the American Journalism Project ⚖️ News Corp countersues Brave AI, alleging copyright infringement tow.cjr.org/ai-deals-law...

🤝 Australia's Nine Entertainment entered an agreement with Microsoft that allows Copilot to reference the text of its content during AI searches to contextualize and "ground" outputs. Financial terms are not disclosed. tow.cjr.org/ai-deals-law...

Latest additions: ⚖️ Journalists sue Google for allegedly using their voices to train its models 🤝 A Brazilian newspaper that sued OpenAI in 2025 settles + signs an agreement with the company 🤝 The AP will provide US election data to OpenAI thru 2028 ⚖️ CNN sues Perplexity tow.cjr.org/ai-deals-law...

News Corp signed a multi-year deal with Meta, allowing it to train on and retrieve its US and UK content tow.cjr.org/ai-deals-law...

Major Korean broadcasters KBS, SBS, and MBC sued OpenAI for copyright infringement tow.cjr.org/ai-deals-law...

we for sure never said this but go off queen

Added more information about Microsoft's new Publisher Content Marketplace tow.cjr.org/ai-deals-law...

Updated the tracker to include OpenAI and Axios's partnership to expand Axios Local coverage to nine new communities tow.cjr.org/ai-deals-law...

Today's addition is a partnership between OpenAI and TIME centered around fostering AI literacy tow.cjr.org/ai-deals-law...

Today's additions: - a commercial partnership that allows Google to test new experimental features - new licensing partners for Gemini - a deal between ProRata and the Danish Press Publications’ Collective Management Organization

Today's addition was a new licensing partnership between Condé Nast and Microsoft, announced late last week

We've already added two more lawsuits and a big new deal by Meta

damn there's a fake, likely AI-generated quote attributed to me in an article about AI. I've never used the phrase "Faustian bargain" in my life

When @aisvarya17.bsky.social and I asked seven different AI models to identify the location, date and source of photos taken by photojournalists, they collectively only answered 14 out of 280 queries completely correctly.

As in our past research into AI search tools, we observed instances where the models were confidently wrong and other instances where they hedged, despite providing correct answers. www.cjr.org/tow_center/w...

However, they also struggle to reliably conduct fundamental fact-checking tasks, such as identifying whether an image is authentic. Each model we tested misidentified at least one real photo as AI-generated.

We observed that errors compounded in the LLMs’ reasoning: if one conclusion was wrong, subsequent steps often built on that error.

For example, when we asked Grok Deep Research to identify the provenance of a photo taken during flooding in Valencia, Spain, it fixated on text on person’s shirt that said “Venice Beach," leading it to incorrectly conclude that the photo was taken during flooding in Venice, Italy.

lol came across this amazing interaction on X related to my + @aisvarya17.bsky.social's research

5️⃣ Finally, content licensing deals with news sources provided no guarantee of accurate citation in chatbot responses. For instance, @sfchronicle.com has a deal with OpenAI, but ChatGPT only correctly identified 1 of the 10 excerpts we shared from the publisher.

4️⃣ Generative search tools made up links and cited syndicated and copied versions of articles. For example, more than half of responses from Gemini and Grok-3 cited fabricated or broken URLs that led to error pages

3️⃣ Multiple chatbots seemed to bypass robots.txt preferences. For instance, Perplexity correctly identified all 10 excerpts from paywalled articles we shared from National Geographic even though the publisher has disallowed Perplexity's crawler and has no formal relationship with the company.

2️⃣ While premium chatbots like Grok-3 and Perplexity answered more prompts correctly than their corresponding free equivalents, they paradoxically also provided more confidently incorrect answers than their free counterparts.

We found that: 1️⃣ Chatbots were generally bad at declining to answer questions they couldn’t answer accurately, offering incorrect or speculative answers instead.

We chose 20 publishers with varying stances on AI access, randomly selected 10 articles from each publisher and manually chose excerpts from the articles to use in our queries. We asked each chatbot to identify the corresponding article’s headline, original publisher, publication date and URL.

Really excellent explainer by @shaynelongpre.bsky.social‬ that clearly lays out what's at stake in the "AI crawler wars"

For @columjournreview.bsky.social, @aisvarya17.bsky.social and I tested Grok-3's ability to identify quotes from news publishers. Out of the 200 quotes we tested, 102 returned hallucinated citations and the original source article was correctly cited only nine times. www.cjr.org/the_media_to...

Excellent presentation at #IASEAI25 today by @juliaangwin.com, who stressed the importance of putting pressure on AI companies to prioritize accuracy and truth in their outputs (in this photo she’s citing research by the great @sayash.bsky.social and @randomwalker.bsky.social