Wednesday, February 25 · ~45s read

Hey 👋 Here's your signal for today:

🔥 Meta bets $100B on AMD to break Nvidia's grip on AI chips
Meta signed a multi-year deal with AMD for 6 gigawatts of Instinct GPUs — with the first 1GW of the upcoming MI450 deploying in H2 2026. This comes days after Meta also expanded its Nvidia relationship, signaling aggressive dual-sourcing of AI compute at a scale that cements AMD as a serious challenger to Nvidia's data center dominance.

⚡ Pentagon gives Anthropic until Friday: drop the guardrails or get blacklisted
The DoD has threatened to designate Anthropic a "supply chain risk" — forcing every Pentagon contractor to certify they're not using Claude in their workflows. Anthropic is reportedly digging in, refusing to strip Claude's safety restrictions for unrestricted military use. The same week: xAI's Grok got cleared for classified military systems. The contrast is deliberate.

📌 Amazon's AGI lab chief is out less than 2 years in
David Luan — who joined Amazon via the Adept acqui-hire to lead agentic AI — announced on LinkedIn he's leaving at week's end to "cook up something new." It's the latest in a pattern of founder-level departures from big-tech AGI labs, raising real questions about whether these labs can hold onto the talent that makes them matter.

That's your Wednesday signal. See you tomorrow.

— The AI Decoded Team

Get the full weekly deep-dive every Monday → getaidecoded.com/subscribe