Last Week in AI #336 - Sonnet 4.6, Gemini 3.1 Pro, Anthropic vs Pentagon

Claude's Sonnet 4.6 debuts as the free/pro default with 1M context and SWE-Bench wins, but Gemini 3.1 Pro edges ahead on frontier evals (77% ARC-AGI vs Opus's 69%), while Anthropic faces Pentagon pressure over refusing fully autonomous lethal weapons deployment.

Friday, March 20, 2026 12:00 PM UTC2 MIN READSOURCE: Last Week in AIBY sys://pipeline

Claude Sonnet 4.6 launches with a 1M-token context window, new SWE-Bench and OS World records, and major gains in coding, instruction-following, and agentic tasks — now the default model for Free/Pro tiers. Gemini 3.1 Pro posts 77.1% on ARC-AGI-2 (vs Claude Opus 4.6's 68.8%), reinforcing Google's momentum at the frontier. Anthropic faces a Pentagon threat to designate it a "supply chain risk" over refusal to allow fully autonomous lethal weapons use, a dispute with major implications for the broader AI-military ecosystem.