Meta’s “Project OT” aimed to slash up to 60 % of certain teams and shift thousands of workers into AI‑focused roles, but the most aggressive phase was scrapped after the technology fell short of expectations.
Key Takeaways
- Project OT explored cutting the size of some teams by as much as 60 % and was structured in two waves (May 2024 and November 2024).
- Meta ultimately cut 8,000 jobs (≈10 % of staff) and moved 7,000 employees into AI‑centric roles before cancelling the second wave.
- Internal data showed AI agents generated volume, not value: code changes rose 220 % YoY, user‑facing changes only 36 %, while security incidents climbed 40 % and firefighting time rose 70 %.
- Employee sentiment fell from 74 % to 55 % in the half‑year Pulse survey, and 26 staff have sued over alleged AI‑driven layoff targeting.
- Meta will still invest at least $130 billion in AI chips and infrastructure this year.
| Metric | Value | Why it matters |
|---|---|---|
| Project codename | OT (Organization Transformation) | Signals a systematic restructuring around AI. |
| Planned team reduction | Up to 60 % | Would have dramatically reshaped several business units. |
| Actual layoffs (May 2024) | 8,000 jobs (≈10 % of staff) | First concrete impact of the AI‑driven plan. |
| Employees moved to AI roles | 7,000 | Shows the scale of internal redeployment. |
What Project OT Intended to Be
Meta’s internal documents, later reported by Reuters, describe Project OT as a year‑long effort to cut costs, redesign team structures, and concentrate staff on “priority work” such as generating training data for AI models. The plan envisioned “talent‑dense” squads supervising AI agents that would perform routine daily tasks previously handled by thousands of engineers, designers, and product managers.
Two‑Wave Timeline and Its Abrupt Change
The rollout was split into two phases. The first wave, slated for May 2024, would have eliminated low‑performing staff and closed open positions. A second wave, originally set for November 2024, was meant to deepen the cuts. On the night of 19 May, hours before the first wave, Zuckerberg called off the November phase. The next day Meta announced the 8,000‑person reduction and the simultaneous transition of 7,000 workers into AI‑focused teams.
Why the Aggressive Cut Was Shelved
Internal performance metrics revealed a widening gap between activity and value. Chief technology officer Andrew Bosworth disclosed that code changes to internal platforms rose 220 % year‑over‑year, yet changes that reached users as new or upgraded features increased only 36 %. Moreover, unchecked AI agents triggered “large‑scale, disruptive actions” that humans rarely perform, leading to a 40 % rise in major technical and security incidents and a 70 % increase in time spent firefighting those issues.
“the trajectory of the agentic development over at least the last four months hasn’t really accelerated in the way that we expected,” Mark Zuckerberg told staff at a July town‑hall (reported by Engadget).
Employee Reaction and Organizational Shifts
Meta engineers were tasked with creating “software puzzles” that served as training data for the AI agents, a routine many described as rote. A tracking system captured keystrokes and mouse clicks on U.S. devices to teach agents human‑like computer interaction; the program was paused in June. Twenty‑six employees have since sued, alleging that AI‑driven layoff decisions targeted workers on medical leave.
Internal sentiment, measured by the half‑year Pulse survey, fell from 74 % favourable to 55 %. Staff responded to executive posts with elephant emojis—a symbolic protest. Labour‑organising activity accelerated, and unit heads began overseeing smaller “pods” of three to five specialists, each titled “builder,” with middle‑management layers removed.
Meta’s Public Position and Ongoing AI Investment
Meta confirmed Project OT, emphasizing that the 60 % figure applied only to specific scenarios, not the entire company, and that many major units were excluded. The firm reiterated its commitment to “priority work” on newly‑established teams and clarified that performance ratings and promotions remain human‑driven.
Despite the setbacks, Meta plans to spend at least $130 billion on AI chips and infrastructure this year, underscoring a long‑term bet on AI even as it publicly promotes a people‑first narrative in advertising and executive essays.
Why It Matters
Project OT illustrates the tension between ambitious AI‑driven efficiency goals and the practical limits of current technology. The initiative’s partial rollout resulted in a sizable workforce reduction, a surge in security incidents, and a measurable dip in employee morale—factors that could influence how other large tech firms design AI‑assisted restructuring plans. Observers will watch Meta’s next moves, especially whether the promised AI benefits materialize within the three‑to‑six‑month horizon cited by leadership.
Frequently asked questions
What was the original scope of the team cuts under Project OT?
Internal documents explored reducing the size of certain teams by up to 60 %, with the intention of replacing routine work with AI agents.
How many employees were actually laid off, and how many were reassigned?
Meta cut about 8,000 jobs (≈10 % of its staff) and moved roughly 7,000 employees into AI‑related roles during the same week.
Did the plan affect the entire company?
No. Meta stated that several major units were excluded and that the 60 % figure applied only to specific scenarios, not the whole organization.
What internal performance data prompted the cancellation of the second wave?
AI agents generated a 220 % increase in internal code changes but only a 36 % rise in user‑visible features, while security incidents rose 40 % and firefighting time grew 70 %.
What legal actions have arisen from the restructuring?
Twenty‑six employees have filed lawsuits alleging that Meta’s AI systems targeted workers on medical leave for layoffs.
Bottom Line
Meta’s Project OT demonstrated both the allure and the risk of replacing large swaths of human labor with AI agents. While the company proceeded with a 10 % workforce reduction and redeployed thousands to AI roles, technical shortcomings and rising security incidents forced it to abandon the most aggressive cuts, leaving a mixed legacy of cost‑saving ambition, employee unrest, and a continued multi‑billion‑dollar AI investment.
Reporting based on: original source.
Comments
Post a Comment