Zuckerberg admits that Meta's AI is progressing slowly
Mark Zuckerberg, CEO of Meta, made an unexpected confession during an internal meeting. He acknowledged that the company's AI agents are not advancing as quickly as he had hoped. This is a cold shower for those who believed that Meta's restructuring, which included job cuts and a reorientation towards AI-focused teams, would yield immediate results.
Expectations were high, but reality, as always, has its own pace. Zuckerberg also commented that the company's reorganization was not as "clean" as it could have been. There was a miscalculation in the timing of the changes. Meta laid off about 10% of its global workforce and reallocated approximately 7,000 employees to AI teams in May. These decisions generated internal resistance and concerns about employee morale.
The idea was to fund heavy investments in AI infrastructure and prepare the company to reap the benefits of the efficiency provided by the technology.
The unfulfilled promise
Zuckerberg admitted that, looking back, the development of AI agents has not accelerated as expected in the last four months. He was referring to automated systems that perform tasks on behalf of users. When the restructuring began to be planned in January and February, there was exaggerated optimism regarding tools like Claude Code from the startup Anthropic. But so far, the results have not materialized.
Meta is projected to spend up to $145 billion on AI infrastructure this year. This is part of a colossal investment of over $700 billion in technology by industry giants. Zuckerberg hopes that the company will start to see significant benefits from these bets in the next three to six months. In the meantime, patience is the order of the day.
The mouse-tracking dilemma
During the same meeting, Andrew Bosworth, Meta's CTO, addressed a recent incident involving the company's controversial mouse-tracking software. A review indicated that no employee data was included in the AI training. The program, which monitors mouse movements and digital activities of employees to train AI, was paused after the exposure of sensitive data. If resumed, it will be on an "opt-in" basis, meaning employees can choose whether to participate or not.
Comments (0)
Comments are moderated and if they violate our Terms and Conditions of use, the comment will be deleted. Persistence in violation will result in a ban of your account.