Safeguard News

The Pentagon cuts ties with Anthropic

The Pentagon declares Anthropic a "supply chain risk"

Anthropic files a lawsuit against the federal government

A judge grants an injunction against the Department of War

An injunction is granted against the supply chain risk designation

OpenAI signs a contract substantially weaker than Anthropic's requirements

The Department of War signs a deal to use OpenAI models instead

Anthropic's "supply chain risk" designation removed before 2027

The supply chain risk designation is officially issued

An injunction is granted against the supply chain risk designation And survives unblocked by other courts for 6 months

Anthropic's "supply chain risk" designation removed before July

The government offers the same terms to Anthropic as to OpenAI

Resignation letter signed by at least 5 OpenAI employees

Anthropic's contract is publicly available

US company with >100B total valuation cancels some Pentagon contract and uses Anthropic supply chain risk designation as stated justification

OpenAI's new contract is publicly available

The Pentagon invokes the Defense Production Act

Dario Amodei leaves or is removed from Anthropic

Anthropic gives the government unfettered access of their own accord (they "cave in")

The Pentagon designates Anthropic a supply chain risk, AND invokes the Defense Production Act

Autonomous weapon + surveillance Claude safeguards are removed for the Pentagon

Resignation letter signed by at least 5 Anthropic researchers

Will Amazon have to divest / break with Anthropic in 2026?

Anthropic stops advancing AI capabilities

The Pentagon continues to use Anthropic services without the requested changes (autonomous weapon + mass surveillance of americans)

YES

YES

The same version of Fable available to Americans is available in most of the world

Fable is available with safeguards more stringent than previously (even for Americans)

Fable is available to both Americans and non-Americans with distinct safeguards for American and non-American users

Fable is available, but not on the API

YES

YES

YES

YES

🧠⚠️ Misaligned goals — AGI misunderstands human values and acts against our interests.

🕵️‍♂️💣 Used by bad actors — hackers, authoritarian regimes, or terrorists weaponize AGI.

🎮❌ Loss of control — humans can't predict or stop AGI’s actions anymore.

🌳💥 Environmental harm — AGI decisions damage ecosystems or speed up climate change.

💼📉 Job loss & chaos — AGI wipes out jobs, leading to mass instability.

Other

🌐🤝 No global coordination — nations or companies rush ahead without cooperation.

👺🐒 Humans are evil — AGI can only align the universe by destroying us.

🤯🔀 Unintended consequences — AGI does something harmful without meaning to.

🛑🔧 Weak safety protocols — not enough safeguards or oversight before release.

🚀💡 Runaway self-improvement — AGI upgrades itself faster than humans can follow.

NO

Somaliland recognised by less than 30 UN members in 2030 / China doesn't recognise

Somaliland recognised by more than 29 and less than 150 UN members in 2030 / China doesn't recognise

Somaliland recognised by more than 149 UN members in 2030 / China doesn't recognise

NO