so a security researcher joins anthropic, stays for about two months and walks away before his equity vests. then immediately starts appearing in giant ai could kill us all talks. tell me it’s a setup without telling me it’s a setup. they're so fucking afraid of open models.