The direction

Where the AI race is heading

AI competition is moving away from a single frontier race and toward control of the path from model to deployed workflow. That movement is being carried by agentic execution inside existing enterprise systems, by partner-led rollouts that package adoption and governance, and by an infrastructure contest widening from chips to datacenter capacity, financing, and power. What is particular now is how directly these layers are locking together into the operating system of scale.

The move

What actors actually did

Microsoft expands Copilot agent features, OpenAI launches Rosalind Biodefense, Anthropic releases Claude Opus 4.8

This week, Microsoft broadened its AI offerings with several notable actions. On May 26, Copilot Studio introduced computer-using agents, making UI-level automation for websites and desktop apps broadly available. The following day, Windows 365 for Agents entered public preview, enabling Copilot agents to handle workflows in Cloud PCs across older and UI-based systems, under enterprise controls. On May 29, Copilot Health became available in preview for US subscribers, offering a secure platform for combining health records and wearable data.

One major competitor, OpenAI, launched Rosalind Biodefense on May 29, giving vetted developers and US government partners access to GPT-Rosalind for biodefense, public health, and pandemic preparedness. Anthropic released Claude Opus 4.8 on May 28, improving coding, agentic tasks, and reasoning, adding new features, and lowering prices. These moves reflect continued expansion and specialization among leading AI providers.

Other actors also made relevant advances. On May 28, IREN Ltd borrowed $3.6 billion to supply Microsoft with Nvidia GPUs, supporting their data center operations in Texas. Anthropic also expanded in Korea, appointing KiYoung Choi as regional director and opening a Seoul office on May 26.

Sources