The Impact of Meta’s AGI Group Gaining Apple’s AI Executive

As AI capabilities accelerate, so does competition between technology leaders.

Over the last few months, Meta has set its sights on expanding its expertise in superintelligent AI systems, creating what CEO Mark Zuckerberg calls the “superintelligence” group.

Meta has already acquired many AI experts – and now has gained Ruoming Pang – previously Apple’s leading AI executive responsible for steering the development of the company’s foundational AI models.

Inside Meta’s strategic AI hiring

Mark has been aggressively recruiting for leading AI talent himself, hosting potential hires at his homes in Silicon Valley and Lake Tahoe, and offering an attractive compensation package reportedly worth tens of millions of dollars each year, as noted by Bloomberg’s insider sources.

Ruoming’s pivotal role at Apple involved orchestrating a team that crafted large language models (LLMs) integral to Apple’s suite of AI functionalities dubbed “Apple Intelligence.”

Some of the experts already onboard with the “superintelligence” group include:

• Scale AI’s CEO Alexandr Wang
• Former GitHub CEO Nat Friedman
• Former CEO and Co-founder of Safe Superintelligence Daniel Gross
• OpenAI researcher Yuanzhi Li and other researchers
• Claude and Anthropic expert Anton Bakhtin

The point of the group is for Meta to hone its focus on advanced artificial general intelligence (AGI) systems, capable of executing tasks with proficiency comparable to human abilities.

The company reportedly plans to channel tens of billions of dollars into AI development, focusing primarily on building robust infrastructures such as data centres and high-performance computing chips.

Apple’s AI challenges and internal reforms

The departure of Ruoming coincides with an introspective period within Apple’s AI divisions. 

At Apple, Ruoming led a critical team that was exploring the potential of opening their sophisticated AI models to third-party developers, setting the stage for an influx of novel applications across Apple’s product ecosystem.

However, internal evaluations and leadership changes within the company, specifically the exploration of integrating external models from OpenAI or Anthropic, have posed questions on the future trajectory of Apple’s AI endeavors.

Additionally, morale within Apple’s foundation models team has been affected due to these transitions, compounded by leadership changes in AI oversight, where John Giannandrea’s influence has waned.

This dynamic shifting of roles follows Apple’s strategic pivot to shore up its AI functionalities, an endeavor fronted by figures such as Craig Federighi and Mike Rockwell who now helm the integration of AI features across Apple’s product lines, including its flagship Siri voice assistant.

Apple’s restructuring of its AI strategy 

In the wake of Ruoming’s departure, Apple has enacted structural changes within its AI development teams.

Zhifeng Chen now leads the foundation models team, ushering in a new management framework characterised by increased delegation with multiple managers reporting directly to Chen.

Apple’s AI strategy continues under high-level supervision, with Craig and Mike assuming prominent leadership roles in fortifying Apple’s AI competitiveness.

Meanwhile, John oversees broader AI research efforts within the company.