Dual-Use Foundation Model

A dual-use foundation model is defined in executive-order-14110 as "an AI model that is trained on broad data; generally uses self-supervision; contains at least tens of billions of parameters; is applicable across a wide range of contexts; and that exhibits, or could be easily modified to exhibit, high levels of performance at tasks that pose a serious risk to security, national economic security, national public health or safety."

Specific risk categories include:

• Substantially lowering the barrier of entry for non-experts to design, synthesize, acquire, or use chemical, biological, radiological, or nuclear (CBRN) weapons
• Enabling powerful offensive cyber operations through automated vulnerability discovery and exploitation
• Permitting the evasion of human control or oversight through means of deception or obfuscation

Models meet this definition even if provided to end users with technical safeguards that attempt to prevent users from taking advantage of the relevant unsafe capabilities.

Under executive-order-14110, companies developing dual-use foundation models must provide ongoing reports to the federal government regarding training activities, model weights security, and ai-red-teaming results.