lucassz

The researchers don't really seem to remark on how surprising it is that the wiki the agents converged on happened to also publicly log the IPs of all visitors, including OpenAI employees, a feature that almost no website has.

Although maybe we can think of that as a selection effect where both this, and the fact that it was possible to edit pages using GET requests, were due to it being ancient, idiosyncratic wiki software.

Certainly not everyone. I find it far more likely that both subscriptions and API pricing are profitable in their own right. The fact that some people get great value out of their subscription (just like some people get great value out of their car insurance) doesn't mean it's subsidized.

I agree with you here, but what's special about tech is that many of us learned all these terms fully casually while using computers as children and teenagers, which would be much less common for a chemist. That makes programmers see a lot of things as computer literacy that most people have rather than specialized knowledge.

> You can't securely have truly stateless authentication without having massive resources, see the cryto.net link above.

I don't think the cryto.net post really explains why this is true (at least in a way that would be made different by "massive resources").

My guess is that the vast majority of those self-reported cases where relating to a take-home assignment (e.g., copying off a classmate's solution). Even without proctoring, you need to be a lot more brazen to cheat on an in-person exam.

Probably LinkedIn

I think the point is that it looks like Gemini 3 was only tested with the generic "Terminus 2", whereas Codex was tested with the Codex CLI.