🎯 AI Alignment - buckman · Scour

Takes on Automating Alignment 👨‍💻AI Coding

lesswrong.com·6d

Frontier Agents: The Next Evolution of AI Applications 🕵️AI Agents

Title: I built a reward analysis tool for AI alignment — here's why reward hacking is harder to detect than you think 🛡️AI Security

dev.to·9h·DEV

Less human AI agents, please 🎼Agent Orchestration

gridthegrey.com·5d·DEV

Import AI 454: Automating alignment research; safety study of a Chinese model; HiFloat4 🛡️AI Safety Evals

importai.substack.com·6d·Substack

Third Symposium on AIT & ML: AI Safety Applications 🛡️AI Safety Evals

lesswrong.com·1d

Diary of a "Doomer": 12+ years arguing about AI risk (part 3: the LLM era) 🤨AI Criticism

lesswrong.com·19h

Alignment Faking Replication and Chain-of-Thought Monitoring Extensions 🐛Fuzzing

lesswrong.com·1h

AI Content Filter Bypass 2026 — How Researchers Test Safety Filtering Systems 🛡️LLM Security

dev.to·5d·DEV

How could I best use this opportunity? (AI Safety) 🛡️AI Safety Evals

lesswrong.com·6h

What holds AI safety together? Co-authorship networks from 200 papers 🛡️AI Safety

lesswrong.com·2d

Anthropic's New Update on Designing AI: How Claude Is Being Built for the Future 🎭Anthropic Claude

dev.to·4d·DEV

Substrate-Sensitivity 🔓Side-Channel Attacks

lesswrong.com·1d

GPT-5.4-Cyber: OpenAI's Game-Changer for AI Security and Defensive AI 🧠OpenAI

dev.to·6d·DEV

The Changing North Star of AI Control 🏛Sovereign AI Infrastructure

lesswrong.com·4d

Monday AI Radar #22 ✍️Prompt Engineering

lesswrong.com·5d

Pando: A Controlled Benchmark for Interpretability Methods 🎯AI Reliability

lesswrong.com·5d

Why AI Safety should be for-profit? 🛡️AI Safety

lesswrong.com·5d

vLLM-Lens: Fast Interpretability Tooling That Scales to Trillion-Parameter Models 🤖LLM Inference

lesswrong.com·3d

A taxonomy of barriers to trading with early misaligned AIs 📊Polymarket

lesswrong.com·5d

No more posts from buckman's subscribed feeds.

Scour all 24391 feeds Learn more about Feeds

Log in to enable infinite scrolling