Claude 365

What It Covers

Anthropic sues the U.S. Department of Defense over its supply chain risk designation, Microsoft launches Copilot CoWork using Claude technology, UK hyperscaler NScale raises $2B at a $14.6B valuation, Gulf AI investment faces war disruption risk, and a 1,488-person study identifies "AI brain fry" as a measurable workplace condition.

Key Questions Answered

• •AI Legal Precedent: Anthropic's lawsuit against the DoD argues that the supply chain risk designation — typically reserved for foreign cybersecurity threats — unconstitutionally penalizes a U.S. company for its AI safety positions. The General Services Administration has already terminated its one-gov contract, cutting Anthropic access across all three federal government branches.
• •Enterprise Agentic AI: Microsoft's Copilot CoWork integrates Claude CoWork technology into Microsoft 365, enabling AI agents to execute multi-step tasks across Outlook, Teams, Excel, and SharePoint simultaneously. Unlike Claude's desktop-based, folder-level sandboxing model, Copilot CoWork operates in the cloud with enterprise-wide data graph access and full IT audit trails.
• •AI Infrastructure Funding Risk: The U.S.-led war in Iran threatens over $300B in planned Gulf AI investments. Drone strikes on Amazon data centers in the region have already elevated project risk. OpenAI and xAI carry greater exposure than Anthropic, having raised Gulf capital and committed to building regional data centers, while Anthropic has avoided large-scale facilities there.
• •Cognitive Overload Threshold: A study of 1,488 U.S. workers finds productivity increases when using one to three AI tools simultaneously, but drops after three. Workers managing high AI oversight levels report 14% more mental effort, 12% more mental fatigue, and 19% greater information overload — concrete signals that agent-heavy workflows require deliberate cognitive load management.
• •Burnout vs. Brain Fry Distinction: AI reduces burnout scores when replacing repetitive tasks, but does not reduce mental fatigue scores under the same conditions. Firms measuring performance by token consumption or AI-generated code lines — as Meta does for engineers — risk accelerating cognitive exhaustion by incentivizing the exact high-oversight, high-workload patterns that drive AI brain fry.