On May 7, 2026, Cloudflare reported a quarter that most companies would celebrate: $639.8 million in revenue, up 34% year-over-year, beating analyst estimates. Non-GAAP net income hit $94 million, up 61%. Free cash flow surged 59% to $84.1 million.
Then they fired 1,100 people — 20% of the entire company.
The AI Restructuring Wave
Cloudflare's CEO Matthew Prince didn't frame the cuts as downsizing. He called it a structural transformation, stating that AI is driving a fundamental re-platforming of the Internet and a paradigm shift in how software is created and consumed.
The numbers tell the story. Cloudflare's internal AI usage increased by over 600% in just three months. Employees across engineering, HR, finance, and marketing now run thousands of AI agent operations daily. Smaller teams are handling work that previously required significantly larger groups.
The company expects $140-150 million in restructuring costs — $105-110 million in cash severance and $35-40 million in stock-based charges. Most of this hits in Q2, with completion expected by end of Q3 2026.
Not Just Cloudflare
On the same day, Freshworks announced it was cutting 500 jobs — 11% of its workforce. The company's CEO Dennis Woodside noted that more than half of Freshworks' code is now written by AI. This is the company's third major round of layoffs in recent years, following a 13% cut in November 2024.
Two major SaaS companies, same week, same reason: AI made their teams too large for the work that remained.
Wall Street didn't reward it either. Cloudflare stock dropped 18% after hours despite the revenue beat. Investors are nervous about short-term restructuring costs, even when the long-term logic is clear.
What This Means for Hosting Customers
When a hosting and infrastructure company cuts 20% of its workforce, the impact flows downstream to customers. Here's what to watch for:
- Support response times may increase. Fewer support engineers means longer queues. If your production environment goes down at 3 AM, will an AI agent understand the nuance of your specific setup?
- Feature development may slow or shift. Fewer product managers and engineers mean fewer features shipping — or features increasingly shaped by what AI can build rather than what customers actually need.
- Documentation gaps widen. Technical writers are often among the first roles automated. When docs don't cover your edge case, troubleshooting becomes harder.
- Account management becomes impersonal. Smaller customer success teams mean less proactive outreach and relationship-building, especially for mid-market customers.
The Human Support Question
There's a fundamental tension in the hosting industry right now. AI agents are genuinely capable — they can write code, debug configurations, monitor infrastructure, and respond to tickets. But hosting isn't just about automated responses.
Hosting is about trust. When your e-commerce store crashes on Black Friday, when a database migration goes sideways, when a zero-day vulnerability needs immediate patching — these moments require judgment, context, and accountability that AI agents don't yet provide reliably.
At DeployBase, we believe the right approach is using AI to augment human teams, not replace them. Our infrastructure automation handles deployment, scaling, and monitoring. But when something goes wrong, you talk to a person who understands your stack and your business.
The Bigger Picture
Cloudflare and Freshworks are signals, not anomalies. The pattern is clear:
- Revenue keeps growing — AI makes products better and cheaper to build
- Headcount keeps shrinking — AI makes teams more productive, reducing the need for large workforces
- Customer experience becomes a gamble — the quality gap between AI-supported and AI-only service will define the next era of hosting
If you're choosing a hosting provider in 2026, ask this question: when things break (and they will), who picks up the phone? An AI agent running on the same infrastructure that's having the problem — or a human engineer who can think outside the runbook?
The companies that get the AI-human balance right will win the next decade of hosting. The ones that cut too deep will learn an expensive lesson when their biggest customers walk.
