Mark Zuckerberg of Meta Mark Zuckerberg shared his vision On Wednesday for “personal superintelligence”, the idea that people can use AI to achieve their personal goals.
Smuggled into the letter is a signal that Meta changes as it plans to release AI models while chasing “superintelligence.”
“We believe that the benefits of superintelligence must be shared with the world as widely as possible,” wrote Zuckerberg. “This said, superintelligence will raise new security concerns. We will have to be rigorous about mitigating these risks and careful about what we choose to open a source.”
This word order for open source is significant. Zuckerberg has historically positioned Meta’s llama family of open models as the company’s key differentiator of competitors like Openai, Xai and Google Deepmind. The purpose of Meta was to create open AI models that were as good or better than those closed models. In a a 2024 letterZuckerberg wrote, “Starting next year, we expect future liame models to become the most advanced in the industry.”
And while many say that Llama does not match the strict definition of an open source AI-partly because Meta did not publish their mass training data-Zuckerberg’s words note a possible change of priority: an open source may no longer be the default for Meta’s cut AI.
There is a reason why Meta’s rivals keep their models closed. Closed models give companies more control over monading their products. Zuckerberg noted last year that Meta’s business is not dependent on selling access to AI models, so “release Llamon does not underestimate our income, sustainability or ability to invest in research as it does for closed providers.” Meta, of course, makes most of his money to sell online advertising.
However, that stated point of view of open models was before Meta began to feel it falls behind competitors, and executives have become obsessed with beating Openai’s GPT 4 While developing llama 3.
Techcrunch -Event
San -Francisco
|
27-29 October 2025
Cut to June 2025, when Meta began his public acting -Sprint seriously by investing $ 14.3 billion in Skala AI,, Getting founder of scale and general manager, and restructuring their AI efforts Under a new unit called Meta Superintelligence Labs. Meta spent billions of dollars Get researchers and engineers from top AI companies and build new data centers.
Recent reports indicate That all this investment led to Meta pause testing on its latest lame model, Behemoth, and instead focus efforts to develop a closed model.
With Zuckerberg’s mission to introduce “personal superintelligence” to the world – a determined change of the rivals he says, works on “automation of every valuable work” – his AI -Money strategy makes it clear. From Zuckerberg’s words today it is clear that Meta plans to deliver “personal superintelligence” with its own products as glasses of increased reality and headphones of virtual reality.
“Personal devices like glasses that understand our context, because they can see what we see, hear what we hear, and interact with us during the day will become our main computer devices,” Zuckerberg wrote in Wednesday’s letter.
When asked about Meta, perhaps holding his most advanced models closed, a Meta spokesman said the company remains compromised with an open source AI and said it was also expected to train closed source models in the future.
“Our position on open source AI is unchanged,” a spokesman said. “We plan to continue releasing major open source models. We have not released everything we have developed historically and we expect to continue training a mix of open and closed models ahead.”
