At an ex-OpenAI researcher's influential lab, $500,000 salaries aren't enough to fix a talent 'bottleneck'
THE SO WHAT
METR struggling to hire even with $500,000 salaries shows that frontier AI safety and evaluation talent is now as constrained as top model research. If you're depending on external labs or a few key hires for assurance, assume long lead times and start building internal eval and red-teaming muscle instead of waiting for a cavalry that may not arrive.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIDisrupting a Criminal Scam Operation
AI-native fraud is now organized enough that labs are running counter-intel ops against specific scam shops. If your product touches payments, dating, or gambling, assume you’re already in an arms race with LLM-augmented fraud rings and budget for continuous abuse tooling, not one-off rules.
Applied AIWho Governs the AI Boom?
Regulation is likely to fragment by risk class — bioweapons and systemic harms on one track, labor and market impacts on another — which means your AI roadmap will straddle very different rule sets. Operators should map their use cases against these buckets now, because compliance, documentation, and audit expectations will not be uniform across the portfolio.
Applied AIAI Adoption Sparks Cultural Divide
When AI becomes a fault line in identity and workplace norms, deployment risk shifts from pure tech and legal into culture and retention. Leaders rolling out AI at scale need a change-management plan — explicit norms, opt-outs, and feedback loops — or they'll end up with shadow resistance that quietly kills adoption.
Applied AIMinnesotas nudification ban takes effect after judge rejects xAIs bid to pause it
State‑level bans on AI nudification show how fast jurisdictional guardrails can harden around specific generative use cases. If you touch image generation or manipulation, you now need a state‑by‑state legal map and technical controls that can enforce it at the feature level.