BREAKING OpenAI has launched GPT-5.4, described as "our most capable and efficient frontier model for professional work."
The release includes two variants:
Key features include state-of-the-art coding, computer use capability, tool search, and 1M-token context window.
GOVERNMENT The US Department of Defense has officially designated Anthropic as a supply-chain risk—making it the first American AI company to receive this label.
Paradoxically, the DOD continues to use Anthropic's AI for operations in Iran, despite the label. This creates a strange situation where the military finds the technology useful but the government won't formally certify it.
Dario Amodei's response: The Anthropic CEO reportedly called OpenAI's messaging around their military deal "straight up lies"—highlighting the growing rift between AI companies over defense contracts.
ENTERPRISE Amazon has launched Amazon Connect Health, an AI agent platform specifically designed for healthcare providers.
The platform handles:
This represents a major push by AWS into vertical AI solutions, targeting the $4.5 trillion healthcare market.
TECHNOLOGY Nvidia's Blackwell architecture has set a new STAC-AI record for LLM inference in financial services.
The record demonstrates Blackwell's ability to handle complex financial computations—particularly relevant for algorithmic trading, risk assessment, and quantitative analysis.
Jensen Huang's twist: Meanwhile, Huang stated Nvidia's investments in OpenAI and Anthropic will likely be its "last"—raising questions about the future of Nvidia's AI partnerships.
PRIVACY Meta is being sued over privacy concerns with its AI smart glasses, after investigations revealed subcontractors reviewed footage including nudity and sensitive content.
Lawyers argue Meta's marketing promised privacy and user control, but the reality involved human reviewers accessing customer footage—raising serious questions about the "AI" labeling of such products.
ENTERTAINMENT Netflix has acquired InterPositive, Ben Affleck's AI-powered filmmaking company.
Unlike generative AI that creates synthetic actors, InterPositive helps production teams work with actual footage for post-production edits—a practical application of AI in the filmmaking workflow.
DEVELOPER TOOLS Cursor (from Anysphere) is rolling out "Automations"—a new agentic coding system that automatically launches AI agents triggered by:
This represents the next evolution of AI coding assistants: proactive agents that work continuously without constant human prompting.
OPEN SOURCE Microsoft released Phi-4-reasoning-vision-15B, a compact multimodal reasoning model available on HuggingFace and GitHub.
Key differentiator: Achieves competitive accuracy with 10x less compute than comparable models, making it viable for deployment on modest hardware.
| Story | Key Point |
|---|---|
| US Chip Controls | New sweeping export controls being considered for AI chips |
| Luma Agents | Creative AI agents with "Unified Intelligence" for text/image/video/audio |
| Railway $100M | AI-native cloud infrastructure startup raises to challenge AWS |
| Lio | $30M Series A for AI-powered enterprise procurement automation |