Good press is helpful, but public investors also care about regulatory risk

"Didn't sell its soul" is a pretty high bar for any large AI lab taking government and enterprise money

Nott really, the dispute is that Anthropic wanted to keep restrictions against domestic mass surveillance and fully autonomous weapons, while the Pentagon reportedly wanted the models available for any "lawful" use

It probably boosts their reputation with one segment of the market while making them much less attractive to another (just my thoughts)

And one of those segments is about 50% of the US population, and the other is about 50% of the US population + the rest of the world.

Source: A Norwegian that just cancelled his ChatGPT plus subscription and will consider Gemini or Claude instead.

This feels bad for the industry. If every AI company learns that having explicit red lines gets you blacklisted, the incentive is to keep safety language vague and negotiable

I suspect that's why experienced officers sometimes intervene like in the OP's story

Clearance forms are weird in that they're not just legal documents, they're inputs into an investigative process

A lot of that comes down to what's objectively verifiable vs what's discretionary, and also what's culturally normalized inside the org.

The danger isn't just being risky, it's being anomalous

So something uncomfortable about clearance processes: they're not purely about truth, they're about interpretable truth