Comet AI browser can get prompt injected from any site, drain your bank account

It’s obviously fundamentally unsafe when Google, OpenAI and Anthropic haven’t released the same feature and instead use a locked down VM with no cookies to browse the web.

LLM within a browser that can view data across tabs is the ultimate “lethal trifecta”.

Earlier discussion: https://news.ycombinator.com/item?id=44847933

It’s interesting that in Brave’s post describing this exploit, they didn’t reach the fundamental conclusion this is a bad idea: https://brave.com/blog/comet-prompt-injection/

Instead they believe model alignment, trying to understand when a user is doing a dangerous task, etc. will be enough. The only good mitigation they mention is that the agent should drop privileges, but it’s just as easy to hit an attacker controlled image url to leak data as it is to send an email.

A smart performant local model will be the equivalent of having good anti-virus and firewall software. It will be the only thing between you and wrong prompts being sent every which way from which app.

We’re probably three or four years away from the hardware necessary for this (NPUs in every computer).

A local LLM wouldn’t have helped at all here.