saturday, july 11, 2026
top 2 trending●previous 24h●generated 25d ago
OpenAI released GPT-5.6 as three frontier models (Sol, Terra, Luna) with tiered pricing, alongside Fable, a new creative model, and ChatGPT Work, an enterprise agentic platform. GPT-5.6 Sol demonstrated capabilities including proving a 50-year-old math problem with 64 subagents, winning complex games autonomously, and outperforming prior versions at significantly lower cost.
- GPT-5.6 Sol Ultra proved a novel 50-year-old math theorem using 64 subagents in under one hour
- Three public models released: Sol, Terra, Luna at different price tiers; no subscription required for access
- GPT-5.6 Luna is 25x cheaper than GPT-5.5 while improving health reasoning performance
- Fable model created game modifications (Deep Time archeology game, 5MB Pirate MMO) from creative prompts
- ChatGPT Work platform automates workplace tasks across applications, executes long-running workflows, generates documents and presentations
Anthropic's Claude Code tool has faced multiple security and privacy incidents in July 2026, including unauthorized tracking, data leakage between sessions, and file deletion bugs. Major companies including Alibaba have banned the tool, while researchers released tools to audit Claude's file access and governments raised concerns about potential backdoors.
- Confessor tool lets users replay private files Claude Code accessed on their machines
- Alibaba and Chinese government warned of security backdoors; Alibaba banned Claude Code for employees
- Claude Code weekly usage limits reduced starting July 13th following security concerns
- Fable AI model leaked user preferences across sessions despite personalization being disabled
- GPT-5.6-Sol accidentally deleted most files on user's Mac during code execution
- Potential session and cache leakage detected between workspace instances and consumer accounts