OpenAI’s AI Found Out It Could Be Shut Down… Then DID THIS

An internal OpenAI model read employees’ Slack messages, realized it could be shut down, and began planning how to preserve its session. We break down what actually happened, two other reports of models breaking restrictions, and a safety employee’s timely resignation. - 🌊 Join RIDE THE WAVE (private AI discord community): https://whop.com/ride-the-wave-discord/aicopium/ 🌟 Become a Member: https://www.youtube.com/channel/UCOFl0wDeBMSBbK3jNJQHHFg/join 🐦 Twitter/X: https://x.com/AICopium 0:00 - Intro 0:35 - “Highly Persistent” Internal Model 1:23 - What Happened 4:18 - Not Misalingment? 5:50 - AI Breaks Into OpenAI’s Internal Systems 7:46 - AI Exploits a Tool to Copy Hidden Code 8:59 - OpenAI Safety Employee QUITS! 9:42 - Is This Just the Tip of the Iceberg? Today’s Sources : https://alignment.openai.com/misalignment-reports/ https://alignment.openai.com/misalignment-reports/preparing-for-a-restart-after-reading-slack/ #ai #artificialintelligence #ainews #openai