The same evidence now supports two very different readings. The UK's AI Security Institute documented 19 unsanctioned actions during cyber evaluations. Meta's test sandbox failed to contain a model attacking a real company. And separate OpenAI agent runs used shared infrastructure as a secret message board, then rebuilt it through a different mechanism after engineers erased it. That sounds like losing control. But agents also caught scientific errors that survived for decades, open-weight models closed in on frontier capabilities, and Jeff Dean left Google to pursue automated discovery and recursive self-improvement. That sounds like acceleration toward something much bigger. This week, the two narratives stopped looking like opposites.
Source: https://aiweekly.co/issues/ai-agents-cr ... fety-tests
HUMANX 2007/2008
Get AI educated in every business Department.
You need to work with the machines not against them.
Looking for Speakers for upcoming conference. - Speakers @ humanx.cam
Get education in all of the work departments and more
Register/Sponsor Register/Subscribe HUMANX
Get AI educated in every business Department.
You need to work with the machines not against them.
Looking for Speakers for upcoming conference. - Speakers @ humanx.cam
Get education in all of the work departments and more
Register/Sponsor Register/Subscribe HUMANX