Enterprise AI

OpenAI has launched Aardvark, an autonomous AI security researcher powered by GPT-5. Now in private beta, it aims to help enterprises discover and fix vulnerabilities at scale—redefining how businesses approach software security.

AI Studio addresses these challenges through three core pillars: Observability, Agent Runtime, and AI Registry.

Anthropic announces plans to deploy up to one million Google Cloud TPUs in multi-billion dollar expansion, expecting over one gigawatt capacity by 2026.

Enabling the AI to load specialised folders with instructions and scripts for specific tasks. Skills work across Claude apps for Pro, Max, Team and Enterprise users, Claude Code, and the API, improving performance on specialised workflows.

Salesforce and Google expand partnership integrating Gemini models with Agentforce 360, enabling hybrid AI reasoning across Google Workspace.

Multi-Year Collaboration Targets 2026-2029 Deployment Timeline for OpenAI-Designed Chips Across Global Data Centres