The context window has been shattered: Subquadratic debuts a 12M token window
(news.ycombinator.com)
571.
573.
Using Claude Code: The unreasonable effectiveness of HTML
(news.ycombinator.com)
574.
575.
Google broke reCAPTCHA for de-googled Android users
(news.ycombinator.com)
576.
Google Broke reCAPTCHA for De-Googled Android Users
(news.ycombinator.com)
577.
Teaching Claude Why
(news.ycombinator.com)
578.
579.
580.
581.
Show HN: Git for AI Agents
(news.ycombinator.com)
582.
Google Cloud Fraud Defence is just WEI repackaged
(news.ycombinator.com)
583.
Human typing habits and token counts
(news.ycombinator.com)
584.
585.
586.
587.
Why Pinterest is telling you to ditch the app and go to a rave
(feeds.feedburner.com)
588.
589.
How Sakana trained a 7B model to orchestrate GPT, Claude and Gemini LLMs
(venturebeat.com)
590.
AI slop is killing online communities
(news.ycombinator.com)
591.
AI Slop Is Killing Online Communities
(news.ycombinator.com)
592.
Natural Language Autoencoders: Turning Claude's Thoughts into Text
(news.ycombinator.com)
593.
Hardening Firefox with Claude Mythos Preview
(news.ycombinator.com)
594.
595.
596.
AI-Driven Cyberattack on Mexico Couldn't Breach OT Systems
(darkreading.com)
597.
'TrustFall' Convention Exposes Claude Code Execution Risk
(darkreading.com)
598.
World's First AI-Driven Cyberattack Couldn't Breach OT Systems
(darkreading.com)
599.
600.