Mark Zuckerberg and Meta are pouring billions into acquiring top AI talent to catch up in the generative AI race, raising questions about whether the spending spree will pay off. OpenAI CEO Sam Altman recently pointed out that Meta is offering massive bonuses—reportedly up to $100 million—to lure engineers, with generous salaries also on the table. In mid-June, Meta invested over $14 billion for a 49% stake in Scale AI, a company that specializes in data labeling to train AI models for clients including governments and businesses. As part of the deal, Scale AI CEO Alexandr Wang joined Meta to contribute to its “superintelligence” initiative.

A Meta spokesperson confirmed the strategic partnership, stating the collaboration would strengthen efforts to generate AI training data and advance superintelligence research. Reports suggest Meta has attempted to recruit other major AI figures, such as OpenAI co-founder Ilya Sutskever, as well as teams from Google competitor Perplexity AI and video AI startup Runway. Zuckerberg is believed to be leading the charge personally amid concerns that Meta is falling behind rivals in AI innovation.

Despite launching the latest version of its Llama model, Meta's performance in AI benchmarks—particularly in coding tasks—has lagged behind top competitors. The company has since formed a new unit focused on developing AI systems that surpass human-level thinking and understanding. Tech analyst Zvi Moshowitz commented that while Meta may attract elite talent, the effectiveness of its strategy remains questionable. He warned of the risks in relying on "mercenary" hires and building products that fail to inspire internal enthusiasm. “I don’t expect it to work, but maybe Llama won’t be as bad,” he remarked.

While Meta’s stock is approaching record highs and the company is valued near $2 trillion, some investors are becoming uneasy. Baird strategist Ted Mortonson noted institutional concerns over how Meta is managing its cash reserves, pointing out a lack of financial oversight with Zuckerberg having significant control over decisions. The company hopes to eventually use AI to supercharge its advertising business by automating everything from ad creation to targeting—potentially reducing the need for ad agencies.

According to CFRA analyst Angelo Zino, hiring AI talent is a long-term play and won’t boost profits immediately. However, investing heavily now is crucial to remain competitive in the next phase of generative AI. The New York Times has reported that Zuckerberg may even consider moving beyond Meta’s in-house Llama model, possibly adopting alternatives. Penn State professor Mehmet Canayaz believes that Meta doesn’t necessarily need the best language model to succeed. With AI agents tailored to specific tasks, Meta can still perform strongly within niche markets.