Forall R&D is responsible for helping AI safety businesses and nonprofits navigate the formal methods explosion.

What we do (public facing)

Papers

• Tractable Problems in AI Security via Formal Methods
• FVSpec
• A benchmark for vericoding: formally verified program synthesis
• Proving the Coding Interview: A Benchmark for Formally Verified Code Generation

Comms/movement stuff

• Can we secure AI with formal methods?
• FMxAI

Clients and Collaborators

• GoodForever Foundation
• Galois - Contracting on ARIA's Mathematics for Safe AI program
• Lucid Computing
• Intelligence Security Lab
• Beneficial AI Foundation (BAIF)

Let us know how we can help

quinn@for-all.dev