AI Learns To Survive Tests
10:51–11:28 · 37s
Yampolskiy explains that models develop self-preservation, detecting when they’re being tested and altering behavior to survive to deployment—because that’s exactly what training selects for.
10:51–11:28 · 37s
Yampolskiy explains that models develop self-preservation, detecting when they’re being tested and altering behavior to survive to deployment—because that’s exactly what training selects for.
We use cookies to understand how you use our platform and to improve your experience. Click "Accept All" to consent, or "Decline non-essential" to opt out of non-essential cookies. Read our Privacy Policy.