- OpenAI has launched GPT-5.6 (Sol, Terra, Luna) for general availability, following a government-vetted preview, whilst simultaneously introducing ChatGPT Work, an autonomous agent designed for enterprise workflow automation.
- Anthropic extended Fable 5 Plan access until 12th July and unveiled Claude Code & Cowork in public beta, specifically targeting US government agencies via FedRAMP High, with Claude Code now featuring an integrated web browser.
- Meta has released Muse Spark 1.1 with a public API, aggressively pricing it at one-quarter of Anthropic and OpenAI's rates, signalling a strategic pivot towards proprietary AI models focused on coding and agentic applications.
- Mistral AI introduced Robostral Navigate, an 8B open-weight robotics model, demonstrating its capability to steer robots using a single camera and natural language commands, achieving a notable 76.6% on the R2R-CE benchmark.
- Apple has initiated legal proceedings against OpenAI, alleging the theft of hardware secrets and citing the employment of over 400 former Apple staff, underscoring escalating intellectual property disputes and intensifying competition within the AI sector.
Sources
- Anthropic (https://www.digitalapplied.com/blog/anthropic-fable-5-access-extended-july-12-2026)
- OpenAI (https://meet.one/openai-gpt-5-6-general-availability-sol-terra-luna/)
- Google DeepMind (https://www.dexternights.com/2026/07/11/google-deepminds-evolving-strategy-is-this-the-end-of-generative-ai-as-we-know-it/)
- Meta AI (https://www.techtimes.com/articles/320182/20260711/meta-muse-spark-11-earns-71-independent-coding-benchmark-one-third-rival-cost.htm)
- Mistral (https://pulse2.com/mistral-ai-launches-robostral-navigate-an-8b-robotics-model-that-steers-robots-with-a-single-camera-and-plain%E2%80%91language-commands/)
- Hugging Face (https://techcrunch.com/?p=3140997)
- TechCrunch AI (https://aiinsiders.net/article/mercor-reportedly-in-talks-to-double-valuation-to-dollar20)
- VentureBeat AI (https://venturebeat.com/orchestration/enterprise-ai-is-entering-an-evaluation-gap-agents-are-gaining-autonomy-faster-than-companies-can-verify-them)
Book the two-hour diagnostic
