An overview of existing deals between news publishers and AI companies and the questions they raise regarding long-term costs, the selection process, and more
That makes the cluster of deals more than a new revenue story: it begins to establish private reference points for what AI companies will pay, which publishers qualify, and what rights are exchanged.
First-order effects
News Corp, the Financial Times, and Dotdash Meredith gain a negotiated channel to monetize content for AI use, while their AI counterparts gain contractually sourced material rather than relying solely on open-web access.
The agreements put deal terms—such as duration, permitted uses, compensation structure, and access to current material—at the center of publishers’ AI strategy.
Second-order effects
Publishers without agreements face pressure to decide whether to pursue bilateral licensing, withhold access, or seek collective leverage; the lack of transparent benchmarks makes those choices harder.
AI companies must manage a more fragmented supply market, where premium publishers can negotiate individually and content access may differ by partner and use case.
Third-order effects
If bilateral deals proliferate, news content is likely to become a differentiated AI input rather than an undifferentiated web resource, with distribution and bargaining power concentrated among large publishers and major model providers.
The unresolved selection and pricing questions could push the market toward more standardized licensing practices or collective and legal mechanisms, though the available coverage does not establish which route will prevail.
The trend: AI developers and publishers are moving from unpriced web scraping toward negotiated, selective commercialization of high-value news content.
Really glad to see by former colleagues at — @TowCenter tracking the flow of money from AI companies to news organizations. This is a really critical story at an urgent moment. (From Pete Brown): — https://www.cjr.org/... And be sure to check out their “AI Partnership Trac…
For those interested in the #AIJournalism deals, we have launched a database of deals @TowCenter and our research director @beteprown has written an analysis of the early deals and questions that remain. A ‘pricing model is emerging’ ...https://www.cjr.org/...
A criticism by @CJR is that this is another instance of big tech picking winners among journalists who are then beholden to them until the cash dries up. Even if you believe that, I'm not sure what the remedy is. Ban deal making for AI training data? https://www.cjr.org/...
New from @TowCenter - our latest Platforms and Publishers work , this time on AI deals with newsrooms. We have created a database, and Research Director @beteprown has written an analysis here https://www.cjr.org/... nice thread below from @_FelixSimon_
Here's a great visual of content deals by AI companies by @CJR. It's somewhat incomplete as content deals with user generated content sites like Reddit, Stack Overflow & Tumblr are not covered. It's hard to imagine how any startup can compete with this level of spending. [image]
I wrote earlier that without transparency around the conditions of the AI licensing deals between tech companies & publishers it is difficult to assess their implications. The @TowCenter's Pete Brown has launched a database providing us with an overview https://petebrown.quarto.p…
One thought: If OpenAI is partially paying publishers in credits to use OpenAI, and Microsoft is funding OpenAI with credits to use Azure... Credits all the way down.
Winner-takes-most. On AI licencing, “the big question for news organizations - or some might argue, the chosen few that are given a choice - is: Deal or no deal? (For those not invited to the table, it's more a case of: Deal with it.)” Pete Brown writes. https://www.cjr.org/...