Fastly Shifts Strategy to Prioritize Edge Security and Compute
Fastly is pivoting to a profitable edge cloud model, reporting record margins and growth in security and compute services at the 2026 Citi TMT conference.
By Muhamed Porić
September 19, 2026 at 2:36 AM

Fastly is pivoting from its traditional content delivery focus toward a more profitable, diversified edge cloud platform. This shift is underscored by surging growth in its security and compute business segments. The company has moved away from a growth-at-all-costs model, as it now reports consistent profitability and record margins.
"Fastly is positioned as cloud-neutral, which gives us an edge in AI inference and agentic AI applications," said Rich Wong, CFO of Fastly, during the 2026 Citi Global TMT Conference.
Financial Performance and Segment Growth
Fastly reported $183 million in revenue for its most recent quarter, a 23.3% increase compared to the same period last year. The company achieved a 65.8% gross margin, reflecting the higher profitability of its newer service offerings compared to traditional bandwidth delivery.
Performance metrics across the company's business units show a divergence in growth:
- Security: Revenue grew 43% year-over-year to $42.1 million.
- Compute and Other Services: Revenue grew 69% year-over-year to $7.3 million.
This growth in high-margin services has contributed to a streak of financial stability. The company has recorded four consecutive quarters of operating profit and six consecutive quarters of positive free cash flow, according to the conference transcript.
Expanding Into AI Infrastructure
To capitalize on the demand for AI-ready infrastructure, Fastly recently launched ContentGuard. The product provides publishers with granular control over how AI agents and bots access, scrape, and monetize their content. This addresses a concern among content creators regarding large language model training data.
By positioning itself as a cloud-neutral provider, Fastly aims to integrate its edge capabilities into enterprise environments that utilize multiple cloud providers. This approach lowers latency for AI inference tasks, which require processing power closer to the end user than centralized data centers typically allow.
What Is at Stake
For Fastly, the transition tests whether it can maintain its market share in content delivery while scaling newer, more specialized services. As enterprises deploy agentic AI, which consists of autonomous programs that execute tasks on behalf of users, the demand for secure, low-latency edge processing is expected to rise. The company's ability to capture this market depends on its success in scaling compute services without diluting the gross margins it has recently established.
Muhamed Porić
Founder and Editor of Embers.
Newsletter
Get Embers in your inbox
The stories that actually moved something, delivered when there's something worth sending, not daily filler.