Between 2020 and 2025, Meta spent over $50 billion on Reality Labs, its virtual and augmented reality division. During that same period, OpenAI launched GPT-4, Google shipped Gemini, and Anthropic built Claude into a serious competitor. The argument that Zuckerberg's metaverse fixation cost Meta its shot at AI leadership is gaining traction among prominent tech investors.
The timing is hard to ignore. Meta rebranded from Facebook in October 2021, signaling VR as its future. That same year, OpenAI was quietly scaling up the research that would produce ChatGPT fourteen months later. By the time ChatGPT hit 100 million users in January 2023, Meta was still pouring billions into Quest headsets and Horizon Worlds - a virtual world that never found a real audience.
Meta has tried to course-correct. Its LLaMA series of open-source models (large language models released for anyone to use and modify) gave it credibility in the AI research community. Meta AI is now embedded across Instagram, WhatsApp, and Facebook. But in the consumer AI assistant race - the one generating actual revenue - Meta trails OpenAI, Google, and arguably Anthropic.
The counterargument: Meta's AI research division existed before the VR pivot and continued producing strong work throughout. LLaMA 3 is competitive with closed-source models on many tasks. Meta's advantage is distribution - billions of existing users across its apps. No other AI company has that.
But distribution only matters if the product is good enough to retain users. And building a best-in-class AI assistant requires focus, talent, and capital that were being absorbed by Reality Labs. The question isn't whether Meta can compete in AI now. It clearly can. The question is whether it could have been leading instead of chasing.