Pang joined Meta in mid-2025 after leaving Apple, where he focused on AI model infrastructure.

At Meta, Pang oversaw key systems supporting next-generation AI models. His exit comes after months of recruitment by OpenAI, and sources said he left Meta last week.

Meta and OpenAI did not immediately respond to requests for comment.

The development reveals the scale of investment companies are making in AI talent. Reports indicate his compensation at Meta was more than $200 million over multiple years.

His work at Meta was a major one in developing the company’s next-generation AI systems, part of its long-term strategy to compete with firms including OpenAI, Anthropic, Apple, and Google.

For OpenAI, Ruoming Pang strengthens its infrastructure and ability to scale. His expertise will support the training of larger multimodal models and more advanced AI workflows.

The hire occurs as governments strengthen review of AI development. Canada, for example, has recently urged OpenAI to increase safety measures or face regulatory consequences.

The tech giant, which owns Facebook, Instagram, and WhatsApp, reported revenue of $51.24 billion for the third quarter of 2025, a 26% increase from a year earlier and above Wall Street’s expectations. 

But costs rose faster, climbing 32%, and a $15.93 billion charge linked to U.S. President Donald Trump’s “Big Beautiful Bill” slashed net income to $2.71 billion. Without that charge, Meta’s adjusted earnings would have been $18.64 billion.

Despite the strong revenue growth, Meta’s shares dropped 8% in after-hours trading as investors reacted to CEO Mark Zuckerberg’s plans to scale up spending on AI infrastructure, a move that could pressure profit margins in the near term.

Zuckerberg told analysts that Meta intends to “aggressively front-load building capacity,” arguing that it’s the right strategy to be ready for faster-than-expected breakthroughs in AI.

“There’s a range of timelines for when people think that we’re going to get superintelligence,” he said. “I think that it’s the right strategy to aggressively front-load building capacity, so that way we’re prepared for the most optimistic cases.”

The company has launched an initiative called Meta Superintelligence Labs, formed in June, to drive its AI goals. Zuckerberg said the unit already has “the highest talent density in the industry,” and Meta is among the top buyers of Nvidia’s sought-after AI chips. “We’re also building what we expect to be an industry-leading amount of compute,” he added.

Meta company now expects to spend between $70 billion and $72 billion this year, up from an earlier estimate of $66 billion to $72 billion, and says next year’s capital expenditure will be “notably larger.” 

Much of this increase will go toward expanding data centre capacity and hiring AI specialists, according to CFO Susan Li, who confirmed employee compensation will be a key cost driver in 2026.

Meta’s aggressive approach comes as Big Tech firms like Microsoft, Alphabet, Amazon, and OpenAI are all scaling up their compute capabilities to support advanced AI models. 

OpenAI’s CEO, Sam Altman, said earlier this week that he hopes the company will one day be able to “add 1 gigawatt of compute every week,” and that could cost upwards of $40 billion per gigawatt.

This ballooning investment across the tech sector has led to talks of a possible “AI bubble,” with analysts warning that spending may outpace returns. “Meta’s earnings reveal the growing tension between the company’s massive AI infrastructure investments and investor expectations for near-term returns,” said Jesse Cohen, senior analyst at Investing.com.

Still, Meta’s core business stays strong. Its AI-driven advertising systems are performing well, with tools that automate campaigns, create persona-based images, and enhance video ad quality for over 3.5 billion daily users across its platforms. 

The company’s expanding ad offerings on WhatsApp and Threads, and sustained growth in Instagram Reels, have strengthened its competitiveness with the likes of TikTok, YouTube Shorts, and Elon Musk’s X.

“While everyone else is still pitching AI moonshots, Meta has quietly turned AI into margin,” said Jeremy Goldman, senior director at Emarketer. “Its ad tools are sharper, its targeting smarter, and its short-form video business is finally paying off.”

For the fourth quarter, Meta expects revenue between $56 billion and $59 billion, slightly above analysts’ projections.

This is tied to Musk’s ongoing lawsuit against OpenAI, in which he accuses the company of abandoning its founding mission.

Court filings made public on Thursday show that OpenAI subpoenaed Meta in June, demanding records related to discussions of Musk’s unsolicited bid. 

According to OpenAI’s lawyers, evidence reveals Musk and Zuckerberg exchanged messages regarding financing or investment support for the deal through Musk’s AI venture, xAI. OpenAI rejected the offer earlier this year.

Meta resisted the subpoena in July and has now asked the court to block OpenAI’s request, arguing that Musk and xAI themselves should supply any relevant documents. 

A Meta spokesperson pointed to OpenAI’s own filing, which stated that neither Meta nor Zuckerberg signed Musk’s letter of intent to acquire the company. Meta has declined further comment.

The issue traces back to OpenAI’s restructuring into a public benefit corporation, a change Musk insists breaches the company’s original non-profit mission. 

OpenAI, however, says the move is necessary to raise funding and preserve its influence in the industry. The court has already rejected Musk’s call for an injunction against the restructuring, ruling that he had not demonstrated sufficient grounds for success.

Behind the case is an increasingly fierce rivalry in the artificial intelligence sector. Meta, which has been struggling to keep pace with OpenAI and Anthropic, has poured billions into its AI ambitions. 

In 2025 alone, the company invested $14 billion in Scale AI and launched Meta Superintelligence Labs to build systems aimed at surpassing human intelligence.

The company has also aggressively recruited talent from its competitors. In mid-2025, at least eight senior researchers left OpenAI for Meta, including Shengjia Zhao, a co-creator of ChatGPT now heading Meta Superintelligence Labs. 

Others such as Lucas Beyer, Alexander Kolesnikov, and Xiaohua Zhai joined from OpenAI’s Zurich office, all previously linked with Google DeepMind. Reports show Zuckerberg has offered compensation packages exceeding $100 million to lure top AI scientists.

Meta’s goal, however, has faced setbacks. Its flagship models underperformed earlier this year, forcing Zuckerberg to personally intervene with a new strategy. 

Court documents from a separate case revealed that Meta executives were fixated on building a system stronger than OpenAI’s GPT-4 but fell short of that goal.

The intersection of Musk and Zuckerberg recalls their history of public clashes, including Musk’s challenge to a cage fight that never materialised. The filings suggest the two billionaires may have set differences aside in pursuit of a common aim: challenging OpenAI’s dominance.

OpenAI has accused Musk of mounting “a relentless harassment campaign” designed to disrupt its operations and strengthen xAI’s position.