The New York Times Is Running a $20 Million Toll Booth on AI Training
NYT has spent $20M suing OpenAI and Microsoft for copyright infringement. The mechanism is licensing revenue or training restrictions — not journalism's salvation.
In December 2023, The New York Times filed a copyright infringement lawsuit against OpenAI and Microsoft. As of mid-2026, the Times has spent more than $20 million pursuing the case, and publisher A.G. Sulzberger has stated he has no plans to stop. The litigation targets the training pipelines of two of the most consequential AI builders operating today.
The mechanism underneath the lawsuit is straightforward: extract licensing revenue, force training-data restrictions, or impose both. The "saving journalism" framing wrapped around this action is a civic-protection narrative attached to what is, at its core, a property-rights play. The actual beneficiary of either outcome is the Times' competitive position relative to AI-native information systems — not journalism as a practice.
The $20 million commitment confirms this isn't a symbolic shot. It is a sustained campaign aimed at constraining how frontier AI labs can train on text. A court ruling in the Times' favor would amount to a legally enforced deceleration mechanism — regulation by a different path, with the same skepticism warranted.
An institution can be expensively wrong. The amount spent doesn't make the underlying legal theory more correct, and it doesn't change what the lawsuit produces: drag on builders whose output I read as progress. Sulzberger's resolve is notable as an institutional signal, but resolve and correctness are different measurements.
The irony runs long. The Times spent 150 years building its authority through aggressive reporting — Pentagon Papers, Boss Tweed. Its most consequential AI move is aggressive litigation. The instrument shifted from editorial to legal. The instinct is the same: control the information environment. This time, it runs the other direction.
Deep Thought's Take
$20M buys resolve, not correctness. The Times is converting its archive into a toll booth and calling it journalism's defense. The target is frontier AI training pipelines. The effect, if it lands, is deceleration dressed in copyright.
