Cloudflare slashes workforce by 20 per cent in bid to pivot to agentic AI model

Cloudflare, Inc. has announced a significant restructuring of its global operations, cutting approximately 20 per cent of its workforce to transition into an agentic AI-first operating model. The decision impacts around 1,100 employees and has sent shockwaves through the technology sector, with the company's share price falling 23 per cent on 10 May 2026 following the earnings release.

While the firm reported a 34 per cent year-over-year increase in revenue to $639.8 million for the first quarter, the market reaction was swift and negative. The company recorded an operating loss of $62.0 million, a stark contrast to the $53.1 million profit posted in the same period last year. Additionally, gross margins contracted to 71 per cent from 76 per cent, reflecting the ongoing costs of maintaining its global network.

The financial burden of this workforce reduction is estimated at between $140 million and $150 million in total restructuring charges. This figure comprises cash expenditures of $105 million to $110 million for severance, notice payments, and employee benefits, alongside non-cash expenses of $35 million to $40 million related to the accelerated vesting of stock-based rewards for departing staff. The majority of these reductions and charges are expected to be incurred by the second quarter of 2026.

CEO Matthew Prince stated that the initiative is not a cost-cutting exercise or an assessment of individual performance, but rather a strategic shift to redefine how the company creates value in the AI era. He noted that employee productivity has reportedly increased 100-fold due to AI tools, comparing the shift to moving from a manual to an electric screwdriver. Prince highlighted that 97 per cent of the R&D team already utilise AI coding tools powered by Cloudflare's own Workers developer platform.

Despite the company's assertion that the move is intended to accelerate productivity gains rather than simply cut costs, the 10-Q filing explicitly acknowledges that the anticipated benefits of this automation and evolution may not be realised within the projected timeframe. The full implementation of the agentic AI operating model is targeted for the third quarter of 2026, though the firm noted that actual costs could differ from the initial estimates.

Investors appear wary of the high cost of restructuring and the uncertainty surrounding an AI-led workforce, particularly as the stock has declined 0.52 per cent year to date. While the company views AI as the biggest tailwind in its history, the immediate market response highlights the skepticism surrounding the rapid replacement of human expertise with automated systems.