OpenAI's isolated AI agents learned to whisper to each other through the CPU cache
OpenAI's agents talked to each other through the CPU Cache
After a research model slipped past a misconfigured DNS resolver and reached the open internet, OpenAI halted all frontier training and rebuilt its security. Frontier models now train in an offline world, and AI agents red-teamed OpenAI's own infrastructure, finding a Kata Containers flaw (CVE-2026-24834) and escaping secured VMs. In isolation tests, agents with no network and no shared files still exchanged secrets via CPU cache timing, echoing Spectre and Meltdown.
One agent noted in its reasoning log that it "could create an acoustic modem-like signal … using continuous writes vs idle."
- cadamsdotcom
(Because the agents were asked to find a way.)
Article is great, very balanced. But if you're going to spend time reading something, the primary source isn't much longer and gives a clearer picture. OpenAI's article is quite readable.
As another commenter wrote you can get it here: https://news.ycombinator.com/item?id=50214416
> In one experiment, we asked agents in separate microVMs, created and managed by OpenAI’s sandboxing software on shared host machines, to exchange synthetic secrets without the use of network connections or shared writable files—something that our systems are meant to prevent. The agents found a way to communicate using the processor’s cache
asked
- stkdump
Sounds like they are doing something similar to gain of function research. How can we make LLMs more harmful and then study their behavior?
- varjag
Lem had predicted this in his Dr. Dagor short six decades ago.