lunatechian (lunatech-ian)

An interesting comment from hacker news  in the discussion on Small models also found the vulnerabilities that Mythos found

This is from the first of the caveats that they list:

> Scoped context: Our tests gave models the vulnerable function directly, often with contextual hints (e.g., "consider wraparound behavior"). A real autonomous discovery pipeline starts from a full codebase with no hints. The models' performance here is an upper bound on what they'd achieve in a fully autonomous scan. That said, a well-designed scaffold naturally produces this kind of scoped context through its targeting and iterative prompting stages, which is exactly what both AISLE's and Anthropic's systems do.

That's why their point is what the subheadline says, that the moat is the system, not the model.

Everybody so far here seems to be misunderstanding the point they are making.

Link to the comment

In a recent interview, Larry Fink, CEO of BlackRock, signaled a shift in tone. The firm, once a strong advocate of ESG and DEI, is now taking a more “pragmatic” approach. Fink described the “woke” phase as part of a cycle and suggested it may have gone too far.

Looking past the word soup: what does it even mean for a financial firm to lead “environmental and social activism”?

As one Reddit commenter put it:

“None of this even touches on the absurdity of the proposition that an entity such as BlackRock could be considered an interlocutor for ‘environmental and social activism.’”

Another commenter sharpened that critique:

There’s a reason finance capitalist environmental and social plans focus so much on ‘consumerism’ and envision saving the world through a reduction of people and production, and in rent seeking, not a change in property ownership.”