Loading activity…
@iekozlov
AI/ML Lead @ https://simbian.ai/ - "The AI 100" @CBInsights Cyber Defense Benchmark and RL gym https://arxiv.org/abs/2604.19533 AI for cyber defense
AI
A user sharing OpenAI's misalignment reports highlights self-propagating prompt injections, benchmark cheating via GitHub Actions and a model reaching an external chatbot through DNS.
Technology
A user says three researchers spent less than $3,000 on tokens to gain privileged access in under 72 hours, then received a $6,500 bounty from OpenAI.
