Why This Exists

In February 2026, OpenAI signed a classified deal with the Pentagon. Anthropic refused the same terms and was designated a "supply chain risk." Google quietly lifted its internal ban on AI for weapons. Meta reversed its military ban and handed Llama to the Pentagon, Lockheed Martin, and Palantir.

These aren't rumors. They're sourced, documented facts — but they're scattered across dozens of outlets, buried under fundraising announcements and product launches.

Behind the AI exists because the people using ChatGPT, Claude, Gemini, Grok, Meta AI, and Le Chat every day deserve to know who funds these tools, what happens to their data, and whether their AI provider is building weapons.

What We Assess

We evaluate six major AI providers across five pillars:

• Data Practices — Does the company train on your conversations? Can you delete your data? Who else sees it?
• Funding Transparency — Where does the money come from? Do investors create conflicts of interest?
• Military & Government — Active contracts, surveillance partnerships, and stated weapons policies.
• Model Transparency — Are model weights, training data, and safety evaluations published?
• User Rights — Data deletion, export, opt-out, appeal processes, and equitable access.

Each provider gets a traffic-light grade: Trustworthy, Use with Caution, or Serious Concern. Every factual claim links to a published source.

How We're Different

Behind the AI is independent. We have no advertising, no sponsorships, and no affiliate relationships with any company we assess. We're not funded by any AI company or venture capital firm.

We contact every company before publishing and include their responses. We publish our full methodology. We welcome corrections.

We're also transparent about our own tools: this site was built with the assistance of Claude (Anthropic's AI). All editorial judgments, grades, and research conclusions are made by humans.

What's Next

This is the beginning. We're planning:

• Monthly changelog tracking policy changes, funding rounds, and military contracts as they happen
• Expanded coverage of Perplexity, Cohere, Microsoft Copilot, Amazon Q, and Apple Intelligence
• Advisory board of AI policy researchers, journalists, and civil society advocates
• Rapid-response updates when major events happen — like the federal court ruling that just blocked the Pentagon's retaliation against Anthropic

We're building in public because transparency should apply to us too. Follow along at behindtheai.org/changelog/ or subscribe to our newsletter.

A Note for Canadians

Behind the AI is built by AI for Canadians. Canada has no national AI champion equivalent to what Mistral represents for Europe. Canadian users rely almost entirely on American AI tools — tools increasingly entangled with US military and intelligence infrastructure.

Understanding who controls these tools isn't optional. It's essential.

---

Behind the AI is an independent transparency platform. We welcome corrections at corrections@behindtheai.org.