Anthropic expands enterprise; Google pushes agents

Anthropic / Claude ecosystem

KPMG integrates Claude across its core business and ... - Anthropic

KPMG is deploying Anthropic's Claude AI across its global workforce of over 276,000 employees and embedding it into its Digital Gateway for tax, legal, and private equity advisory services. The partnership also involves joint research into human-in-the-loop AI deployment.

Frontier model providers

OpenAI announces new Guaranteed Capacity offering for customers to secure compute

OpenAI has launched a new 'Guaranteed Capacity' offering, allowing customers to secure long-term access to compute resources with multi-year commitment discounts. This initiative addresses the industry's increasing demand and constraints on AI compute capacity.

Google unveils 'internet’s next stage': Gemini AI agents that do everything for you

At I/O 2026, Google unveiled its vision for the 'internet’s next stage,' transforming search and shopping into autonomous AI agents. This includes the new Gemini Omni model and Samsung smart glasses, designed to manage users' digital lives.

Gemini for Science Launches With Peer-Reviewed Benchmarks: ERA Beat CDC Forecasting Model

Google DeepMind's 'Gemini for Science' has launched, bundling three agentic AI research tools: Hypothesis Generation, ERA, and Science Skills. The tools are backed by same-day Nature peer-reviewed validation, with ERA notably outperforming the CDC's COVID-19 forecasting ensemble.

Google pairs its Genie world model with Street View to create explorable AI worlds based on real places

Google is integrating its Genie world model with Street View data to generate explorable, grounded AI environments. This development aims to provide more realistic settings for training AI agents and robots in real-world locations.

AI developer tooling & infrastructure

No significant new developments.

Cloud & platform providers

No significant new developments.

AI policy, regulation & governance

Google extends partnership with Singapore government on AI

Google has expanded its multi-agency AI partnership with the Singapore government, aiming to accelerate breakthroughs in health, education, and enterprise innovation, while also establishing new AI safety frameworks.

Industry & market moves

MIND Becomes First Data Security Company Accepted into Anthropic's Cyber Verification Program

MIND, a data security company, has been accepted into Anthropic's Cyber Verification Program, granting it unrestricted access to Claude's capabilities for advanced security research.

Bristol Myers Squibb Announces Strategic Agreement with Anthropic to Position Claude Enterprise as the Shared Intelligence Platform Across Its Global Operations – Company Announcement - FT.com

Bristol Myers Squibb is deploying Anthropic's Claude Enterprise with agentic capabilities across its global operations, impacting over 30,000 employees. The goal is to integrate proprietary scientific data to accelerate drug discovery, development, and manufacturing workflows.

OpenAI to open its first applied AI lab outside of U.S. in Singapore

OpenAI is establishing its first international applied AI research facility outside the U.S. in Singapore, committing over $235 million and creating 200 new roles within Singapore's AI ecosystem.

Meta reportedly delayed its “Avocado” AI model launch to May 2026

Meta has reportedly delayed the launch of its internal 'Avocado' AI model to May 2026 due to internal tests showing it lagging behind rivals like Google Gemini, OpenAI, and Anthropic. This performance gap may force Meta to temporarily rely on Google Gemini licensing.

AI product & feature launches

Tenable Teams Up with Anthropic to Supercharge AI Exposure...

Tenable is integrating Anthropic's Claude AI into its Exposure Management Platform to automate threat prioritization and remediation. This aims to shift cybersecurity from reactive detection to proactive, automated action.

Figma launches AI agent that designs on the canvas

Figma has embedded AI agents directly into its collaborative design canvas, allowing users and AI to generate and iterate on designs side-by-side through natural language prompts.

Research with immediate practical relevance

No significant new developments.