OpenAI o3-mini-high mode released: significantly enhanced math, science, and complex coding reasoning
OpenAI rolled out o3-mini-high reasoning mode with major improvements in competition-level math (AIME, IMO), scientific derivation, and multi-step complex programming, featuring longer chains of thought and self-correction for research and high-difficulty engineering use cases.
new
Meta Llama 4 early test version leaked: native multimodal + tool calling + ~1.5M token context
xAI Grok-3 advances in real-time tool use and multi-agent collaboration for parallel complex tasks
OpenAI o3-mini-high mode released: significantly enhanced math, science, and complex coding reasoning
Google DeepMind Gemini 2.5 Pro leads in long-context multimodal tasks with native 1M token window for whole-book reasoning
Altruist Hazel AI tax planning tool generates personalized strategies in minutes, disrupting traditional wealth management
Doximity integrates PeerCheck AI into DoxGPT for real-time peer validation and diagnostic support
Agentic AI in healthcare and life sciences marketing projected to unlock $450 billion value by 2028
Anthropic Claude Opus 4.6 uncovers over 500 zero-day vulnerabilities in open-source code with minimal prompting
Goldman Sachs tests autonomous AI agents for process-heavy work; Intuit, Uber, State Farm trial
Databricks CEO: SaaS isn't dead, but AI will soon make it irrelevant