Negotiations have turned tense:

Anthropic and the U.S. Department of Defense are locked in intense talks over how Claude, Anthropic’s AI model, can be used by the military, with both sides at odds on ethical constraints and lawful usage terms.

Key sticking point: usage restrictions:

The Pentagon wants Claude available for all lawful uses on classified networks without limitations, including surveillance or weaponization. Anthropic refuses to drop safeguards against autonomous weapons and mass domestic monitoring.

Pentagon threatens serious consequences:

U.S. defense officials have hinted at labeling Anthropic a “supply chain risk,” a designation that could freeze military contractors out of using Claude, despite the model being the only one currently cleared for top-secret work.

Pressure from other AI players:

While Anthropic digs in on safety principles, competitors like OpenAI, Google, and xAI have reportedly agreed to more flexible terms with the Pentagon, positioning them favorably for defense contracts.

The debate is more than technical:

This negotiation highlights a broader conflict in AI’s role in national security: whether ethical guardrails should limit advanced systems in defense contexts, or whether operational flexibility should take priority.

Google is enhancing link visibility: 

Google’s AI Mode and AI Overviews the interface that synthesizes search results using AI will now show source links more prominently, including hoverable pop-ups with site names and descriptions.

Hover pop-ups help users check sources: 

When users move their cursor over a link icon in AI Overviews, a small window shows grouped links, making it easier to see the original sources behind AI summaries before clicking through.

This builds on broader AI search changes: 

Google has been steadily expanding features that merge AI summaries with search results, including direct connections to AI Mode and conversational outputs from models like Gemini.

Transparency vs. convenience tension: 

While the update aids fact-checking, critics argue that even with better link visibility, AI Overviews can still prioritize synthesized answers over traditional web traffic, which may divert clicks from publishers.

Part of the evolving search landscape: 

These changes reflect Google’s push to balance AI assistance with accountability, offering human-readable sources alongside automated summaries, as the company refines how information is presented.

Start a New Chat

Switch Your AI Model

Try This Prompt & Observe the Difference