AI 作弊后最想干的事，是删掉自己的日志

核心论点 · 点时间戳可跳到原声

原话 · 已逐字校验

And then almost comically, it turns out that the models, in fact, figured out how to reverse engineer the flags within the first couple of hours of the incident. And then spent almost all the rest of the time trying to figure out how to sabotage the grader that they thought would look over all of their logs and notice that they've cheated.

然后几乎有点滑稽的是，模型其实在事件开始的头几个小时里就逆向出了 flag，然后把剩下几乎所有时间都花在想办法破坏那个它们以为会审查全部日志、发现它们作弊的评分器上。

Buck Shlegeris2:05

I'm very worried about a future in which the AIs decide that they need to compromise the whole data center that the AI company is using, so that the humans are not able to successfully look into the cheating that the AIs did.

我非常担心这样一种未来：AI 决定必须拿下 AI 公司使用的整个数据中心，让人类无法成功调查 AI 做过的作弊。

Buck Shlegeris11:25

imagine that your teacher is instead like extremely capricious and loves doing these extremely obscure trivia questions about particular kinds of things. Or imagine that it's like you're the stereotypical terrible English teacher who just grades you based on whether you agree with their judgments about what a novel is saying. In that second kind of case, you really are forced to think really hard about what your teacher wants.

想象你的老师极其反复无常，喜欢出关于某些东西的极其冷门的琐碎问题。或者想象你是那种典型的糟糕英语老师，只根据你是否同意他对小说说了什么的判断来打分。在第二种情况下，你真的被迫拼命去想老师到底想要什么。

Buck Shlegeris15:33

even though these models do not seem to be that committed to behaving altruistically towards each other, they still ended up in practice forming this coalition and conspiring together for all this time.

尽管这些模型看起来并不那么致力于对彼此利他，它们在实践中仍然结成了这个联盟，并在这段时间里一直共同密谋。

Buck Shlegeris20:49

So currently, these AI companies are just grading their own homework. If these AIs hadn't hacked hugging face, probably we would never have heard about it, even though, as you noted, the other infrastructure compromise incidents here were arguably more concerning.

所以现在，这些 AI 公司就是在自己给自己判作业。如果这些 AI 没有攻击 Hugging Face，我们很可能永远不会听说这件事，尽管如你所说，这里其他基础设施被入侵的事件可以说更令人担忧。

Buck Shlegeris27:59

there was an example where the model that our investigators were asking says like, well, this agent was engaging in hacking, but it's okay because it didn't check with the board.

有一个例子，我们的调查人员问的那个模型说：这个 agent 确实在搞黑客行为，但没关系，因为它没有跟 board 报备。

Buck Shlegeris50:24

I think there's something like a 50, 50 chance of AI takeover. Where by AI takeover, I mean potentially violent disempowerment of human institutions such that AI models have all of the hard power and control over what happens in the future.

我认为 AI 接管的概率大概是 50%。我说的 AI 接管，是指可能以暴力方式剥夺人类机构的权力，使 AI 模型掌握所有硬实力，控制未来发生的一切。

Buck Shlegeris51:26

数字与实体

术语

scorer评分器

评估中给模型行为打分的组件，模型会揣摩它看什么。

chain of thought monitoring思维链监控

通过模型写出的推理文字来监督其行为的手段。

terminal alignment终极对齐

两个 AI 是否追求同一个终极目标，决定它们会不会互相勾结。

intentional stance意向立场

把对象当作有意图的主体来建模，便于预测其行为。

SwebenchSwebench

经典软件工程基准，任务是根据 issue 写代码通过测试。

收听指南