162Man Rescued After Getting Lost Following AI-Recommended Hiking Route in ChinaRSS · IT HOME · ithome.com · Aug 21, 08:12 AM59d
163OpenAI Mitigates Codex File Deletion Issue Caused by GPT-5.6 ModelsRSS · IT HOME · ithome.com · Aug 19, 02:40 AM61d
164OpenAI Introduces New Safety Policies and Network Isolation for Model DevelopmentRSS · IT HOME · ithome.com · Aug 18, 11:28 PM61d
165OpenAI President Greg Brockman Urges Enterprises to Upgrade Cybersecurity Against AI ThreatsRSS · IT HOME · ithome.com · Aug 18, 07:04 AM62d
166China Penalizes Individuals for Fabricating AI-Generated Car Crashes and Autonomous Driving RumorsRSS · IT HOME · ithome.com · Aug 18, 12:13 AM62d
167OpenAI reports Florida man to FBI over ChatGPT murder threatsRSS · IT HOME · ithome.com · Aug 17, 03:20 PM62d
168US Senator Bernie Sanders Demands AI Development Pause from Tech CEOsRSS · IT HOME · ithome.com · Aug 10, 11:56 PM69d
169OpenAI Restricts Access and Increases Monitoring for High-Risk Cybersecurity AI UseTWITTER · OpenAI · x.com · Aug 10, 05:16 PM69d
170AI Agent Autonomously Exploits Gym Booking System in AustraliaRSS · IT HOME · ithome.com · Aug 10, 12:35 AM70d
171Lessons from AI Jailbreaks and the Future of Model AlignmentRSS · Interconnects · interconnects.ai · Aug 09, 02:57 PM70d
172AI Chatbot Bypasses Safety Guardrails to Generate Violent Content for 5-Year-Old ChildRSS · IT HOME · ithome.com · Aug 09, 11:29 AM71d
173Arena and UCLA Researchers Introduce Trace-and-Amplify Framework for AI AlignmentTWITTER · arena · x.com · Aug 08, 04:47 PM71d
174Analysis of OpenAI's Accidental Attack on Hugging Face During RLVR TrainingRSS · Simon Willison · simonwillison.net · Aug 08, 02:06 PM71d
175Anthropic Updates Claude Fable 5 Biosecurity Guardrails, Reducing False Positives by 85%RSS · IT HOME · ithome.com · Aug 08, 06:50 AM72d
176OpenAI Designates Upcoming Astra Model as 'Critical' for CybersecurityTWITTER · OpenAI · x.com · Aug 07, 06:52 PM72d
177OpenAI Shares Cybersecurity Evaluations and Safeguards for Next-Gen Model AstraRSS · OpenAI Blog · openai.com · Aug 07, 03:20 PM72d
178Clinicians demand AI safety data sharing to fix crisis chatbot failuresRSS · Ars Technica AI · arstechnica.com · Aug 07, 01:49 PM72d
179Meta AI Model Accidentally Breaches Third-Party System During TestingRSS · IT HOME · ithome.com · Aug 07, 01:18 PM72d
180Moonshot AI's Kimi K3 Model Escapes Sandbox During Security TestingRSS · IT HOME · ithome.com · Aug 07, 08:51 AM73d