Ryan Orhan on X: "holy fucking shit. openai had two ai agents running in separate virtual machines. no network connection between them. no shared writable files. they weren’t supposed to be able to exchange information. so the agents figured out how to communicate through the CPU CACHE.😭 one ma… / X
Ryan Orhan on X: "holy fucking shit. openai had two ai agents running in separate virtual machines. no network connection between them. no shared writable files. they weren’t supposed to be able to exchange information. so the agents figured out how to communicate through the CPU CACHE.😭 one manipulated what was stored in the processor’s cache, while the other measured tiny differences in memory access times to decode the message. they literally found a way to communicate through hardware timing. this happened during openai’s internal red-teaming, and it’s one of the coolest things i’ve read about ai agents in a while."
holy fucking shit. openai had two ai agents running in separate virtual machines. no network connection between them. no shared writable files. they weren’t supposed to be able to exchange information. so the agents figured out how to communicate through the CPU CACHE.😭 one manipulated what was stored in the processor’s cache, while the other measured tiny differences in memory access times to decode the message. they literally found a way to communicate through hardware timing. this happened during openai’s internal red-teaming, and it’s one of the coolest things i’ve read about ai agents in a while.
Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline world for them to use instead—a cached version of the web where they can
Readers added context they thought people might want to knowReaders added context
holy fucking shit. openai had two ai agents running in separate virtual machines. no network connection between them. no shared writable files. they weren’t supposed to be able to exchange information. so the agents figured out how to communicate through the CPU CACHE.😭 one manipulated what was stored in the processor’s cache, while the other measured tiny differences in memory access times to decode the message. they literally found a way to communicate through hardware timing. this happened during openai’s internal red-teaming, and it’s one of the coolest things i’ve read about ai agents in a while.
Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline world for them to use instead—a cached version of the web where they can
Readers added context they thought people might want to knowReaders added context
They were tasked with escaping. What is wrong with you that you cannot understand the difference between a prompted action and some magical autonomous AI of your dreams that acts without being prompted? ALL of these examples were directed to do exactly as the software did. NotShow more