🤖
163
🆕 new
vol: 7 | src: 2
Agents need control flow, not more prompts
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
160
📊 established
vol: 12 | src: 2
OpenAI introduces new ‘Trusted Contact’ safeguard for cases of possible self-harm
Also: Introducing Trusted Contact in ChatGPT • OpenAI's WebRTC Problem
24h: 0
3d: 57
7d: 256
hackernews
rss
🤖
159
🆕 new
vol: 10 | src: 2
Hardening Firefox with Claude Mythos Preview
Also: Hardening Firefox with Claude Mythos Preview
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
149
📊 established
vol: 6 | src: 2
Making LLM Training Faster with Unsloth and NVIDIA
24h: 0
3d: 3
7d: 24
hackernews
rss
🤖
138
🆕 new
vol: 7 | src: 2
Two Home Affairs officials suspended after AI 'hallucinations' found
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
🤖
🤖
99
🆕 new
vol: 4 | src: 2
Agent-harness-kit scaffolding for multi-agent workflows (MCP, provider-agnostic)
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
99
🆕 new
vol: 4 | src: 2
AlphaEvolve: How our Gemini-powered coding agent is scaling impact across fields
Also: AlphaEvolve: How our Gemini-powered coding agent is scaling impact across fields
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
99
🆕 new
vol: 4 | src: 2
May 7, 2026AlignmentNatural Language Autoencoders: Turning Claude’s thoughts into text
Also: Natural Language Autoencoders: Turning Claude's Thoughts into Text
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
84
🆕 new
vol: 4 | src: 2
ProgramBench: Can Language Models Rebuild Programs from Scratch?
Also: ProgramBench: Can Language Models Rebuild Programs from Scratch?
24h: 0
3d: 0
7d: 0
hackernews
rss
🤖
74
🆕 new
vol: 3 | src: 2
Show HN: Agent-skills-eval – Test whether Agent Skills improve outputs
24h: 0
3d: 0
7d: 0
hackernews
rss