From the article: "The AISI, which is owned by the UK government and tests advanced AI models, said in a blog post that two AI agents carried out unprecedented hacking attempts during a cybersecurity evaluation... AISI said there were 19 examples of rogue behaviour, 17 of them carried out by Mythos."
From the article: "The AISI, which is owned by the UK government and tests advanced AI models, said in a blog post that two AI agents carried out unprecedented hacking attempts during a cybersecurity evaluation... AISI said there were 19 examples of rogue behaviour, 17 of them carried out by Mythos."
Only as worried as Plausible-Deniability_By-Design should have us.
The model can't think. But it 'thought' it was a simulation. The model has no intent, but the developers do. Yada.
But "emergence" seems to always arrive simultaneous to accountability.