Voice & Audio OpenAI Opens GPT-Live-1.1 to the API and Prices Silence at 2 Cents a Minute Coding Assistants Replit Pitted 32 Claude Agents Against NumPy Tuning: 18.9x Speedup While GPT-6.2 Swarms Collapsed at Scale Security GAO Covert Tests: All Four Top Chatbots Produced Viable Attack Plans in Under 12 Hours Once Guardrails Fell Meta Meta Puts Its 2030 Net-Zero Target Under Review as AI Data Centers Push Net Emissions Up 45 Percent Business Daniel Ek Is Backing Europa Fusion, a Stealth UK Startup Led by a Former DeepMind Researcher Betting on Pulsed Power Coding Assistants Stripe Tested 14 Coding Agents on 69 Payment Scenarios: 78% of Criteria Passed, but Only 29% of Runs Met Every Requirement Security A Transcription Firm Used by Police and Prosecutors Left Interview Transcripts and Body-Cam Footage on an Open S3 Bucket News UN Killer-Robot Talks End With a 2028 Deadline and No Binding Rules. About 100 States Wanted a Treaty; the US, Russia and Israel Blocked It Voice & Audio OpenAI Opens GPT-Live-1.1 to the API and Prices Silence at 2 Cents a Minute Coding Assistants Replit Pitted 32 Claude Agents Against NumPy Tuning: 18.9x Speedup While GPT-6.2 Swarms Collapsed at Scale Security GAO Covert Tests: All Four Top Chatbots Produced Viable Attack Plans in Under 12 Hours Once Guardrails Fell Meta Meta Puts Its 2030 Net-Zero Target Under Review as AI Data Centers Push Net Emissions Up 45 Percent Business Daniel Ek Is Backing Europa Fusion, a Stealth UK Startup Led by a Former DeepMind Researcher Betting on Pulsed Power Coding Assistants Stripe Tested 14 Coding Agents on 69 Payment Scenarios: 78% of Criteria Passed, but Only 29% of Runs Met Every Requirement Security A Transcription Firm Used by Police and Prosecutors Left Interview Transcripts and Body-Cam Footage on an Open S3 Bucket News UN Killer-Robot Talks End With a 2028 Deadline and No Binding Rules. About 100 States Wanted a Treaty; the US, Russia and Israel Blocked It

Luke Metz, an Early OpenAI and ResNet Veteran, Joins Meta's Superintelligence Lab as the Talent War Grinds On

Luke Metz, an Early OpenAI and ResNet Veteran, Joins Meta's Superintelligence Lab as the Talent War Grinds On

Luke Metz, a veteran AI researcher who left OpenAI for Mira Murati's Thinking Machines in 2024 and returned to OpenAI earlier this year, has joined Meta's superintelligence lab this week, according to Axios. He reports to Chief AI Officer Alexandr Wang, continuing Meta's aggressive poaching run.

Luke Metz, one of the quiet constants of the modern deep learning era, has joined Meta's superintelligence lab, according to Axios reporting confirmed by people familiar with the move. Metz arrived this week and reports to Alexandr Wang, Meta's chief AI officer, adding another recognizable name to a roster Meta has been assembling almost monthly since it paid billions for the Scale AI deal and installed Wang to lead its AI push.

Metz's path is a map of the industry's last half-decade. He was a co-author of the ResNet paper — "Deep Residual Learning for Image Recognition," the 2015 architecture that made hundred-layer networks trainable and became one of the most cited works in computer vision — during his time at Google Brain. He later joined OpenAI, left in 2024 for Mira Murati's startup Thinking Machines, returned to OpenAI earlier this year, and now lands at Meta, reporting to a lab that has made recruiting from rivals its most visible strategy.

The move was first flagged publicly in an X post by ArfurGrok, an account that has accurately anticipated a series of recent AI moves, before Axios confirmed it with sources. Neither Meta nor Metz has issued a public statement, and details of his role beyond reporting to Wang are not yet known — a reminder to treat the exact scope of the hire as unconfirmed until Meta says more.

For Meta, the hire is less about a single resume than about the pattern. The superintelligence lab has launched several new models in recent months, and Wang's team keeps pulling senior researchers from OpenAI, Thinking Machines and other rivals, betting that concentrated human talent — not just compute — decides who ships the next frontier system first. Every crossover like Metz's is small by itself; the aggregate is the clearest signal of where the talent war's front line sits.

For the labs losing people, the calculus is harsher. Researchers who move between three competing labs within two years are, in effect, pricing their loyalty at each stop — and the institutions paying top dollar for them are buying incrementally less exclusivity each time. The superintelligence race is often narrated through benchmarks and data-center gigawatts, but its true scarce resource keeps changing employers.

Comments (0)

Log in to join the discussion

Log In

No comments yet