The Boy That Cried Mythos: Verification is Collapsing Trust in Anthropic
Вы, вероятно, слышали в недавних новостях о модели Mythos от Anthropic и о том, насколько она хороша в поиске уязвимостей... Со слов самих Anthropic.
Эксперт по компьютерной безопасности Davi Ottenheimer решил проверить, насколько заявления о возможностях Mythos соответствуют правде.
TL;DR: даже собственный документ Anthropic даёт понять, что их инструмент, мягко говоря, не соответствует заявленным возможностям. Ниже мои отдельные цитаты, но я крайне рекомендую прочитать статью целиком.
I’ve been getting more and more curious about the risk from Anthropic’s Claude Mythos Preview. So I pulled the system card, a whoppingly inefficient 244-page document that devotes just seven pages to the claim that the model is too dangerous to release
The flagship demonstration of “unprecedented cyber capability” is in fact a model that weaponized two bugs that a different Anthropic model had already found, in software Mozilla had already patched, in a harness with the actual defenses turned off, where the “triage” step it performed is also performed by its predecessor.
The bugs Anthropic used to justify a $100 million consortium, eleven Fortune-100 partners, a “too dangerous to release” decision, and global headlines that “frightened the British” — an open-weights 3.6B-parameter model finds them too, for eleven cents per million tokens.
(^проверено в AISLE)
The Mythos system card tested the model against small-scale enterprise networks with no active defenses and the model succeeded. The same document tested the model against a properly configured sandbox with modern patches and the model failed.
Автор также утверждает, что Anthropic на волне хайпа отвоёвывает себе тёплое местечко под солнцем в обход существующих ограничений.
By withholding Mythos from general release and granting access only through the Glasswing consortium — Apple, Google, Microsoft, Amazon, Broadcom, Cisco, CrowdStrike, JPMorganChase, Nvidia, Palo Alto Networks, the Linux Foundation — Anthropic inserts itself as a de facto clearance-granting body for an “uplift” of vulnerability knowledge. Without a statutory basis. Without congressional oversight. Without FOIA exposure. Without a neutral arbiter. With a partner list drawn entirely from the largest incumbents in the industry it claims to be protecting.
<...>
That is not a safety posture. It’s regulatory capture dressed as restraint. And it is being constructed with no democratic input, in a legal vacuum, by a private company whose business model depends on selling access to the very capability it has declared too dangerous to release.