Pentagon drops $800M on AI while boards burn cash on hype

Instead of slow homomorphic encryption, the post revives a 1980s solution, secure multi‑party computation (MPC), that provides perfect secrecy without any computational assumptions and runs orders of magnitude faster. MPC lets us sandbox arbitrarily dangerous AIs while guaranteeing no party learns anything about the code, addressing Paul Christiano’s long‑standing ‘cryptographic boxes’ challenge.

Nik Suresh warns that AI hype is hijacking large firms' decision‑making. Executives, often clueless about the tech, force AI‑centric strategies, leading to costly missteps and cultural chaos.

The DoD’s Chief Digital & AI Office awarded contracts worth up to $200 million to Anthropic, Google, OpenAI and xAI to prototype “agentic” AI for military missions. The four deals could total $800 million, marking a massive pivot toward commercial off‑the‑shelf models rather than traditional defense contractors.

A former Google DeepMind researcher has drafted a concrete policy that draws firm red lines, banning autonomous targeting without human control and untargeted profiling, and sets up an independent Review Body to audit government AI contracts. By forcing yearly transparency reports and limiting corporate loopholes, the framework aims to keep high‑stakes AI deployments accountable and could guide future legislation or corporate governance.