You need to be more specific. OpenAI's commitment to assist the Trump administration with domestic mass surveillance seems to have been largely memory-holed.

You're right, unfortunately. How naive of me to think that at least the HN audience would care.

Yeah, it's eerie, same with how everyone seems to have forgotten that OpenAI betrayed democracy by committing to work on unsupervised autonomous weapons and domestic mass surveillance.

Honestly I find comments like yours much more eerie. By all accounts they never agreed to any of that but you say it with such confidence like it's a fact.

The Trump administration's handling of Anthropic showed that regardless of what the contract or the law says or means, they will severely penalize any vendor who refuses their demands. And OpenAI stepped right into that relationship immediately after the administration showed that. So either they were signing up for a supply-chain risk designation and whatever other punishments the Trump administration dreams up, or they're complying.

If this sounds crazy to you, though, I'd like to know, and understand why. I miss ChatGPT/Codex.

Is it hitting intermediate milestones with solid pre-written and human-reviewed acceptance tests? If not, sounds like a very risky commitment.

I went to a conference and talked to all the vendors about their products.

According to the article, you can fix this by turning off Gemini Apps Activity in https://myactivity.google.com/product/gemini

I already have this turned off. It's slow, but I use a browser extension to save my Gemini chats locally, when I want to keep them.

> In a subsequent round, it generally can't meaningfully introspect on its prior internal state

It has the K/V cache, no?

The K/V Cache is just an optimization. But yeah you would expect the attention for the model producing "Ok im doing X" and you asking "Why did you do X?" be similar. So i don't see a reason why introspection would be impossible. In fact trying to adapt a test skill where the agent would write a new test instead of adapting a new one i asked it why and it gave the reasoning it used. We then adapted the skill to specifically reject that reasoning and then it worked and the agent adapted the existing test instead.

It's quite likely that OpenAI is running a significant PR campaign to compensate for the bad rep they earned by stepping in to meet the demands of the Trump administration, after Anthropic refused to assist the administration with mass domestic surveillance and development of lethal autonomous weapons. Presumably OpenAI didn't buy the podcast TBPN just because they like the guys.

https://paulgraham.com/submarine.html

I think there's definitely more scope for ruling out vulnerabilities by implementing simpler designs and architectures.

They have no values of their own, so you have to direct their attention that way.