SAN FRANCISCO, USA — An incident during a call open to thousands of employees has exposed growing frustration within Meta’s Applied AI team. The unit was formed in March to support AI researchers at Meta Superintelligence Labs. It consists of approximately 6,500 product managers and engineers.
Three current employees spoke to WIRED on the condition of anonymity because they were not authorized to speak to the media. They report widespread dissatisfaction with how the company assembled the group and the specific tasks they have been assigned to improve AI models.
Workplace Dissatisfaction
Some employees describe their current assignments as menial. One employee noted that tasks such as generating puzzles to test the reliability of AI models from Meta and other companies are easy compared to the software development work they performed previously. This employee stated that “almost all” staff members seem unhappy.
“Most people find the work soul-crushing,” a third employee said. The situation highlights a sharp contrast between the expectations of the staff and the reality of the AI projects.
The disconnect exists between the capabilities of the team and the work they are asked to perform.
Leadership Perspectives
Cox stated he needed to reckon with how he and other leaders could “get in touch with the company again” and avoid being “overearnest” about the power of AI. He described the technology by saying, “It is neither god, nor is it the devil.” Cox added that AI “is nowhere near as good as you think it is, and it is nowhere near as bad as you think it is.” He noted that the technology “changes every week … and it doesn’t know what day of the week it is.”
Corporate Response
Meta CEO Mark Zuckerberg addressed the situation in an internal memo on Friday. He acknowledged that recent organizational changes had caused distress across the company. “Given the complexity of these changes, we’ve made mistakes and will almost certainly make more,” Zuckerberg wrote.
He stated that as the company navigates this period, he is focused on providing as much stability going forward as possible. The turmoil within the Applied AI team reflects broader challenges in implementing and developing AI technologies.
To create a more productive and stable environment for developers and researchers, Meta must prioritize the concerns of its employees. The company’s ability to leverage skills more effectively will be a key factor as it moves forward. With the commitment to stability and the acknowledgment of organizational complexities, there is potential for improvement in the development of AI technologies.
Background and why it matters
Meta AI began as Facebook Artificial Intelligence Research, or FAIR, in 2013, when the company was still known simply as Facebook. It was established to pursue long-term research in fields such as computer vision, machine translation, and natural language processing. Yann LeCun, a prominent figure in deep learning, directed the group until 2018, when he was succeeded by Jérôme Pesenti, a former chief technology officer at IBM’s big data unit. The lab was renamed Meta AI after the parent company rebranded as Meta Platforms in 2021.
In plain terms, Meta AI is the company’s research division devoted to artificial intelligence, and it is also the name of the virtual assistant that Meta has built into its social networking apps and hardware devices. The assistant appears in the company’s messaging and social platforms and is pre-installed on Ray-Ban smartglasses and virtual reality headsets. It can process voice commands, answer questions, and generate content.
Over the years, the research group has made widely used contributions to the field. It helped create the open-source machine learning framework PyTorch, which became a standard tool for deep learning projects at many companies and research institutions. In 2016, FAIR joined Google, Amazon, IBM, and Microsoft in forming the Partnership on Artificial Intelligence to Benefit People and Society, an industry effort aimed at guiding responsible development of the technology.
The importance of Meta AI has grown alongside the company’s broader shift toward artificial intelligence. Its virtual assistant now influences the news and advertising people see on Meta’s platforms, and the company has said it will use interactions with the assistant to personalize content and ads. That places the research division close to the center of how Meta makes money and how users experience its products. The recent friction reported by employees, involving the gap between research ambitions and day-to-day tasks, matters because the unit is positioned to shape not only Meta’s future but also the daily experience of billions of users.
{“@context”: “https://schema.org”, “@type”: “NewsArticle”, “headline”: “Meta AI Staff Voice Frustration Over ‘Soul-Crushing’ Tasks”, “description”: “SAN FRANCISCO, USA — An incident during a call open to thousands of employees has exposed growing frustration within Meta’s Applied AI team. The unit”, “datePublished”: “2026-06-11T09:30:00+00:00”, “dateModified”: “2026-08-16T12:00:00+00:00”, “author”: {“@type”: “Person”, “name”: “Sophie King”}, “publisher”: {“@type”: “Organization”, “name”: “Info Pulse Today”, “url”: “https://infopulsetoday.com/”}, “articleSection”: “”Artificial Intelligence””, “mainEntityOfPage”: {“@type”: “WebPage”, “@id”: “https://infopulsetoday.com/tell-him-hes-a-piece-of-sht-metas-new-ai-unit-is-a-total-mess/”}, “image”: “https://infopulsetoday.com/wp-content/uploads/2026/08/trendimg_23494.jpg”}
