Meta Halts Broader AI Workforce Reduction Plan
Meta called off planning for a second, broader workforce reduction hours before its May layoffs, Reuters reported on August 26. Internal documents reviewed by Reuters showed that Project OT had explored cutting some teams by as much as 60% as part of an "AI native" organizational redesign. Meta still laid off 10% of employees in the first wave and reassigned roughly 7,000 workers into AI-focused roles, according to Reuters reporting and an internal memo cited by Fox Business.
Meta canceled planning for a second wave of AI-linked workforce cuts hours before conducting layoffs in May, Reuters reported on August 26. Internal documents reviewed by Reuters showed that the company had explored reducing some teams by as much as 60% under a project code-named Organization Transformation, or Project OT.
Reuters reported that the proposed restructuring was designed around an "AI native" operating model in which software would assume more daily work and smaller groups of employees would oversee virtual workers. Scenario planning contemplated two waves: an initial reduction in May and a further shake-up in November. According to Reuters, a human-resources executive projected cuts that could equal or exceed Meta's roughly 25% workforce reduction three years earlier.
Meta proceeded with the first reduction, laying off 10% of its employees the next day, Reuters reported. But Zuckerberg called off planning for further cuts on the night of May 19, according to the news agency.
May restructuring combined cuts and transfers
A May internal memo from Chief People Officer Janelle Gale, cited by Fox Business, described the transfer of roughly 7,000 employees into AI-focused roles while the company removed management layers and flattened organizational structures. Fox Business reported that the affected teams included Applied AI Engineering and Agent Transformation Accelerator.
Gale wrote in the memo that organizational leaders had incorporated "AI native design principles" into new structures, and that many organizations could operate with flatter structures and smaller groups. Fox Business reported that Meta had nearly 78,000 employees at the end of March, citing securities filings.
The reported changes place workforce design alongside model development and infrastructure spending in Meta's AI program. Reuters Breakingviews noted that Meta's employee compensation, excluding severance costs, rose nearly 30% in the first half of 2026 from the same period in 2025, while the company continued to spend heavily on data centers and AI talent. That commentary also cited LSEG estimates projecting Meta investment near $170 billion next year.
Automation claims meet organizational constraints
Reuters' account identifies a material gap between exploring automation-driven headcount reductions and carrying out a company-wide restructuring. The report said employees resisted the proposed changes and that the planned second wave was canceled before the first layoffs occurred.
For ML and data leaders, the reported episode is a reminder that agent deployment is not equivalent to organizational automation. Companies undertaking comparable transitions generally need to define task boundaries, human review paths, access controls, reliability measures, and ownership before reducing the teams that hold operational knowledge. AI-generated code and workflow agents can increase throughput in bounded settings, but enterprise use also expands the importance of evaluation, observability, incident response, and security review.
The reporting does not establish which internal AI systems, if any, were considered sufficiently capable to replace particular roles. It does show that Meta examined workforce changes at a scale far beyond the May layoffs, then halted additional planning.
Key Points
- 1Reuters reports that Meta explored team reductions of up to 60%, but canceled planning for a second workforce-cutting wave in May.
- 2The May restructuring paired a reported 10% layoff with roughly 7,000 transfers into AI-focused roles, according to an internal memo.
- 3Comparable enterprise agent transitions require evaluation, governance, and operational ownership before automation can reliably substitute for established human workflows.
Scoring Rationale
The reported restructuring concerns a leading spender on data centers and directly links workforce design to internal AI automation. It is highly relevant to enterprise AI leaders, though the reporting does not disclose the specific models or systems used to automate work.
Sources
Public references used for this report.
Practice with real Ad Tech data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Ad Tech problems
