/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

OpenAI's deals with publishers are a hedge against a scenario in which scraping becomes harder, training material more expensive, and real-time data more scarce

This past week, Axel Springer, the German media conglomerate that owns Politico and Business Insider, signed a “multiyear licensing deal” …

New York Magazine John Herrman

Context & Ripple Effects

OpenAI had already tested a content-for-technology exchange through its two-year AP licensing arrangement. The Axel Springer agreement extended that approach to publishers behind Politico and Business Insider, tying licensed archives to planned news-summary features.

The deal is significant because it treats publisher relationships as an operating input for both model development and timely product responses, rather than as a one-off distribution partnership. Subsequent reports of talks with Gannett, News Corp, and IAC indicate the Axel Springer pact was part of a broader outreach effort.

First-order effects

  • OpenAI gains licensed access to Axel Springer material and a basis for ChatGPT news summaries, while Axel Springer gains a commercial channel and access to OpenAI technology.
  • Politico and Business Insider content becomes more directly integrated into OpenAI's licensing strategy, making their publisher owner's terms and editorial assets immediately more valuable to the lab.

Second-order effects

  • Other major publishers gain a clearer reference point for negotiating AI licenses, as OpenAI's outreach broadens beyond the earlier AP content-and-technology deal.
  • AI developers seeking current, high-quality news material face pressure to choose between negotiated access and dependence on web scraping, potentially raising the strategic value of large publisher portfolios.

Third-order effects

  • If such agreements proliferate, licensed content could become a recurring input cost and competitive differentiator for AI products that need reliable, current information.
  • The market may consolidate bargaining power around publishers with distinctive archives or broad news brands, while smaller outlets have less leverage unless licensing frameworks become easier to aggregate.

The trend: Generative-AI companies are shifting from treating web content as freely available training material toward negotiated, commercially managed content supply.

Discussion

  • @jwherrman John Herrman on x
    once again, a group of tech companies are invoking eminent domain over the entire internet https://nymag.com/...
  • @nymag @nymag on x
    The deal OpenAI just inked with German media conglomerate Axel Springer offers a preview of how AI and the media might eventually intersect, writes John Herrman. http://nymag.com/...
  • @jjvincent James Vincent on x
    it's true - my favorite ongoing diary of the AI shakeup/shakedown is being written by @jwherrman https://nymag.com/... [image]
  • @jwherrman John Herrman on x
    it's worth thinking about why a content automation company would be interested in paying to license content https://nymag.com/...
  • @alexgraveley Alex Graveley on x
    2023 AI in a nutshell: It's all about the data. Good performance is on the other side of seeing training data that looks *nearly exactly* like what you're trying to do. Generate, verify, retrain on what works - this is the blueprint. When you plateau, optimize generation. ♻️