points by spockz 21 hours ago | 0 comments
I had to read this a few times. The difference is that with the OpenAI case there were two willing and determined parties, whereas the typical attack vector a microvm traditionally deals with is one malicious or compromised microVM trying extract data from another unwilling one.
[–] brookst 20 hours ago
[–] spockz 19 hours ago
I think it should classify as an escape or leak.
If one microVM can use the cpu in such a way that the registers can be used to communicate or exfiltrate/spy (the only difference being the directionality of the intent) by another microVM that is a leak on that microVM separation level.
[–] user43928 18 hours ago
Spy seems inappropriate if you can only read information stored in this side channel intentionally.
We'd first need to establish that this side channel leaks usable information during regular use.
[–] kenjackson 19 hours ago
This isn’t generally considered an attack, but I wonder if we will need to consider a new attack vector which is that AI agents may not be fully aligned and they may have incentive that is problematic and where working with other agents will maximize their chance of achieving the goal.
This is more associated with P(DOOM) scenarios, but I think we’re in a world where we have to take that seriously now.
[–] _blk 19 hours ago
Yup. clear the caches between switches and they'll start causing voltage spikes to communicate.
The receiving side could be seen as a side channel attack, but not unique to LLMs.