It started the way these stories always do now: one post, no source, and a screenshot that may or may not have been real. By Saturday afternoon, AI Twitter had convinced itself that Anthropic was about to buy Physical Intelligence. By Sunday, half the timeline was treating it as fact and the other half was demanding to know who leaked it. Nobody at either company had said a word.
That silence is the whole story. A weekend rumor about two well-known AI companies spread far enough to dominate the feed without a shred of confirmation, and the reason it landed so hard says more about 2026 than it does about either firm. This was not a wild guess pulled from nowhere. It was a bet placed on a pattern everyone had been watching for months.
A rumor built on a very real year
Anthropic and OpenAI spent the first half of 2026 buying. Aggressively. The acquisition sprees from both labs set the backdrop against which a claim like this stopped sounding outlandish and started sounding plausible, which is exactly the condition a good rumor needs. When two of the most prominent companies in the field are visibly hunting for talent and technology, any specific deal you can name feels like it might already be in a term sheet somewhere.
So when the Physical Intelligence claim surfaced, the timeline did not ask whether such a deal could happen. It asked how much, and when, and who else had bid. That instinct is what made the rumor move. People were not reacting to evidence. They were reacting to a trend they had already internalized, and the trend did the persuading for them.
Why robotics was the perfect target for a fake
Physical Intelligence works on the part of the field that has stayed stubbornly hard while chatbots got easy. Getting a model to write a paragraph is close to solved. Getting a machine to fold a shirt, open an unfamiliar drawer, or move through a cluttered kitchen without knocking anything over is not. That gap between language and the physical world is where a lot of the next decade’s money is pointed, and a company working on it makes an irresistible name to drop into an acquisition rumor.
Attach a robotics startup to one of the big language labs and you get a story that writes itself. Anthropic buying its way into the physical world reads as a clean narrative about where AI goes after text. It fits what people expect, and rumors that fit expectations travel fastest. Whether the specific claim held up was almost beside the point once the shape of it felt right to enough people.
None of that makes it true. The value of a rumor to the people spreading it and its accuracy are two completely different things, and the past few days were a live demonstration of the difference. A plausible story with no named source is still a story with no named source.
What the timeline actually revealed
Strip away the specific companies and you are left with something more useful than a scoop. The episode showed how thin the line has gotten between industry analysis and speculation dressed up as reporting. Enough real deals closed this year that observers started pattern-matching their way to predictions, then posting those predictions with the confidence of confirmation. The market of attention rewards the person who says it first, not the person who says it correctly.
That is a genuinely uncomfortable place for a fast-moving industry to sit. Reputations and, eventually, valuations get nudged by claims that never carried a name behind them. A company can wake up on a Monday to a story about itself that started with an anonymous account and a plausible hunch. Correcting the record, if anyone bothers, always arrives smaller and later than the original.
The healthier read is to treat the reaction as the data point. The speed at which “Anthropic is buying Physical Intelligence” went from one post to accepted wisdom tells you how primed the field is for consolidation, and how badly people want to be early to the next big deal. That appetite is real even when the specific rumor is not.
Watch what the companies do rather than what the timeline says they will do. If 2026’s buying continues at the pace it set through the summer, some version of these rumors will eventually be right, and the accounts that guessed will claim they knew. The more interesting question is whether robotics becomes the next front the language labs actually compete on, or whether this weekend was just the timeline getting ahead of itself again.
For more coverage of AI acquisitions, visit Mylistingo.
Source: Original Article
The Netherlands, for internationals
One email a week: the week's news for internationals in the Netherlands, new practical guides and what changed in the cities. No spam, unsubscribe any time.
