Descriptions:
Craig Hewitt’s weekly AI briefing for the week of March 15, 2026 covers a dense slate of product releases and policy developments across the frontier model landscape. On the model side, the episode notes the near-simultaneous release of GPT-5.4 (OpenAI), Claude Opus 4.6 (Anthropic), Gemini 3.1 Pro (Google), Llama 4 Maverick (Meta), and Grok 420 (xAI) — framing the pace of releases as evidence of intensifying competition that ultimately benefits end users.
Two ecosystem developments get substantial attention. First, Anthropic has been designated a supply chain risk to the U.S. military, effectively removing it from Pentagon contracts; OpenAI has stepped in as the replacement vendor. Anthropic has filed an injunction against the Department of Defense in response. Hewitt frames this as a preview of how frontier AI companies transition from commercial tools to essential government infrastructure — and what that means for commercial users caught in the political crossfire. Second, both Stripe and Ramp have released frameworks for AI agents to make payments autonomously, covering both conventional card rails (Visa, Mastercard, Amex) and crypto. Hewitt argues this is a foundational infrastructure moment for agentic commerce.
Rounding out the episode: Anthropic’s new code review feature inside Claude Code (priced at $25–$50 per use on team and enterprise plans), Google’s official CLI for Workspace enabling agents to interact with Gmail, Calendar, and Drive, and Replit Agent 4’s parallel frontend/backend/infrastructure build capability.
📺 Source: Craig Hewitt · Published March 15, 2026
🏷️ Format: News Analysis
