AI talent – Alexandr Wang says it’s inaccurate to portray Meta’s AI lab hires as purely money-driven, pointing instead to compute, autonomy, and research focus.
Meta’s push to build one of the most formidable AI teams has been accompanied by a familiar storyline: that researchers are being hired like mercenaries. chasing the biggest checks in the so-called AI talent war.. Alexandr Wang. the executive leading Meta’s SuperIntelligence Lab. pushed back on that portrayal. calling it unfair and arguing that the day-to-day reality inside the lab does not match outside assumptions.
Wang made the comments during an interview connected to the “Core Memory” podcast. where he addressed how people outside the company view Meta’s recruitment efforts.. When asked about the reputation that the lab’s researchers were lured from rival companies at least partly through extremely generous terms. Wang said the framing misses what drives most researchers to join.
A key part of the debate centers on the compensation narrative.. Wang was asked about reports that top AI researchers received offers reportedly worth $100 million.. He rejected the idea that money is the primary motivator for most of the recruits. saying that their financial outlooks for staying where they were could already be very strong.
Instead of pay. Wang pointed to practical research advantages that. in his view. matter more to scientists and engineers looking to make progress quickly.. He said many joined because the team had high compute available per researcher. allowing them to move faster and take on research work that might not be possible—or might be slower—in their previous environments.
Wang also described the recruiting pitch as being rooted in a mix of talent concentration and freedom to pursue ambitious directions.. People. he said. were drawn to a “cracked” group that was small but dense with expertise. along with the expectation that Meta would provide resources and room for bold research bets.
The lab’s status as a major prize in the broader AI talent war is reflected in Wang’s own background. He is a cofounder of Scale AI and was widely viewed as a top target. Meta spent $14 billion to acquire nearly half of Scale AI and to bring Wang in to lead a new AI team.
Meta’s hiring effort did not stop with Wang, according to the report.. The lab’s roster also includes Nat Friedman. the former CEO of GitHub. who helps lead the development of AI products.. Ruoming Pang. previously heading Apple’s foundation model team. is also among the key recruits. as is Trapit Bansal. a former top researcher at OpenAI.
Beyond compensation, Wang acknowledged that rival executives have mocked the lengths Meta reportedly went to woo AI researchers.. The interview referenced remarks by OpenAI’s chief research officer Mark Chen. who said Zuckerberg hand-delivered homemade soup to an unnamed OpenAI employee—a moment that underscored how personal Meta’s efforts were intended to feel.
Wang said he did not think the soup was homemade, but he still acknowledged the significance of the gesture.. In his telling. the point wasn’t only the item—it was the message that the company was willing to show how much it cared. and that it wanted to demonstrate attention to specific research directions rather than treating recruiting as a simple transaction.
He tied that approach to the broader premise behind building the lab: Meta had to signal that it valued the technology and the researchers’ particular work. Wang said the recruiting process was individualized, aimed at aligning the researchers’ goals with what the lab would provide.
For readers tracking the business side of AI. the dispute over motives highlights a tension that often shows up during executive-level talent races: public narratives tend to simplify hiring into a contest of compensation. while insiders emphasize operational factors such as compute access and the ability to run experiments at scale.
The emphasis on dedicated compute per researcher also matters beyond any single lab. When high compute becomes the bottleneck, hiring can turn into an arms race not only for people, but for the infrastructure needed to turn ideas into models, experiments, and measurable progress.
Wang’s remarks suggest Meta views its recruiting strategy as more than a competition over salary.. By connecting researchers’ decisions to resources. autonomy. and a concentrated talent environment. Meta is effectively arguing that it is selling a working research platform—one where ambitious projects can be attempted faster than elsewhere. even when compensation is already attractive.
AI talent war Meta AI Alexandr Wang SuperIntelligence Lab dedicated compute Scale AI acquisition