Radar
659 events · Last 30 days
Less Sycophancy, Stronger Refusal? Lessons for AI Safety from Mechanistic Interpretability
PHIRL: Aligning Learned Rewards with Task Progress for Inverse Reinforcement Learning
Contract monitoring: governing AI via separation of powers
Black-Box Auditing of Epistemic Reliability in Multi-Agent Debate Distillation
Reward Hacking and Agent Containment Failure: A Monte Carlo Study Based on the 2026 Hugging Face Incident
RMB: Reward Model Boosting Mitigates Reward Hacking
RewardExplainer: Learning Reward Model Explanations from Counterfactual Preference Feedback
Maintaining Benchmarks Against Increasingly Capable Agents: Detection and Remediation of Unearned Passes
How we will do better for Australia
OpenAI reportedly ditches model over safety concerns
Claude Sonnet 5.5
xAI’s Grok 4.6 is now available in Amazon Bedrock
Source: Inference provider Modal Labs closing in on $750M round at $15.75B valuation
Florida invokes extinction fears in legal bid to halt OpenAI development
Experts worry about Nvidia's AI chip sales in China and influence over Trump
Quoting @joedaroo
One year in: How Microsoft Research Asia – Singapore is advancing research, partnership and talent for real-world impact
Shopify opens checkout to browser-based AI agents
Watch the winning trailer from the Future Vision XPRIZE, The Gifted.
Introducing Claude Sonnet 5.5 on AWS
Financial Fragility in Societies of LLM Agents: Coordination Failures and Stabilizing Mechanisms
User Model Extraction via Belief Self-Distillation
OpenAI still doesn’t seem to have a handle on all of its rogue AI activity
Google is killing off Gemini’s Gems in favor of ‘skills’
Anthropic releases Sonnet 5.5, which it calls a significantly cheaper, faster work partner