alt Hacker NewsMore details on the difference between the OpenAI and Anthropic contracts from one of the Under Secretaries of State:
>The axios article doesn’t have much detail and this is DoW’s decision, not mine. But if the contract defines the guardrails with reference to legal constraints (e.g. mass surveillance in contravention of specific authorities) rather than based on the purely subjective conditions included in Anthropic’s TOS, then yes. This, btw, was a compromise offered to—and rejected by—Anthropic.
https://x.com/UnderSecretaryF/status/2027566426970530135
> For the avoidance of doubt, the OpenAI - @DeptofWar contract flows from the touchstone of “all lawful use” that DoW has rightfully insisted upon & xAI agreed to. But as Sam explained, it references certain existing legal authorities and includes certain mutually agreed upon safety mechanisms. This, again, is a compromise that Anthropic was offered, and rejected.
> Even if the substantive issues are the same there is a huge difference between (1) memorializing specific safety concerns by reference to particular legal and policy authorities, which are products of our constitutional and political system, and (2) insisting upon a set of prudential constraints subject to the interpretation of a private company and CEO. As we have been saying, the question is fundamental—who decides these weighty questions? Approach (1), accepted by OAI, references laws and thus appropriately vests those questions in our democratic system. Approach (2) unacceptably vests those questions in a single unaccountable CEO who would usurp sovereign control of our most sensitive systems.
> It is a great day for both America’s national security and AI leadership that two of our leading labs, OAI and xAI have reached the patriotic and correct answer here
Replies
A government promise that they'll only do lawful things is not reassuring at all:
1. We've seen government lawyers write memos explaining why such-and-such obviously illegal act is legal (see: torture memo). Until challenged, this is basically law.
2. We've seen government change the law to make whatever they want legal (see: patriot act)
3. We've seen courts just interpret laws to make things legal
A contractor doesn't realistically have the power to push back against any of these avenues if they agree to allow anything legal.
(At the risk of triggering Godwin's Law, remember that for the most part the Holocaust was entirely legal - the Nazi's established the necessary authorization. Just to illustrate that when it comes to certain government crimes, the law alone is an insufficient shield.)
You're quoting social media posts from a regime official who says he didn't participate in these negotiations and doesn't work for the relevant department.
If his characterization of the agreement is correct, which I will not believe and you should not believe until a trustworthy news outlet publishes the text, I suppose this would convince me that Hegseth does not literally plan to build a Terminator for democracy-ending purposes. There's a lot of inexcusable stuff here regardless, but perhaps merely boycotting OpenAI and the US military would be a sufficient response if this all checks out.