Voice AI went from research project to $500 million a year in revenue in about two years. That’s not a trend — it’s a market shift.

ElevenLabs CEO Mati Staniszewski confirmed in a CNBC interview on May 5, 2026 that the company has crossed $500 million in annual recurring revenue. The trajectory is striking: the company ended 2025 with around $330 million in ARR, added $100 million in net new ARR during Q1 2026 alone to reach approximately $450 million, then pushed past $500 million in Q2.

How Fast This Market Is Moving

The pace matters as much as the number. ElevenLabs went from $330M to $500M ARR in roughly four months. Most software companies take years to generate that kind of incremental growth.

The acceleration is driven by enterprise adoption. Recent contracts with Deutsche Telekom, Revolut, and Klarna represent companies that are not experimenting with voice AI — they’re deploying it at scale for real customer interactions. Deutsche Telekom operates critical communications infrastructure across Europe. Revolut and Klarna are fintech companies where customer trust and interaction quality are core to the business model.

ElevenLabs completed its $500 million Series D at an $11 billion valuation in February 2026, backed by institutional investors including BlackRock, Wellington Management, D.E. Shaw, Schroders, NVIDIA, and Salesforce. The investor list tells you something too: these are not speculative bets on a technology that might work. They’re infrastructure plays on a category that already works.

What Is Actually Driving the Demand

Voice AI has moved from novelty to what some are calling “voice infrastructure” — the layer that sits between a business’s data and its customers, handling interactions in natural language at scale.

The businesses adopting voice AI at this pace are not doing it to save a little money on support tickets. They’re doing it because the alternative — scaling human support teams at the rate modern businesses need — is operationally impossible for most organisations.

A business that adds 10,000 new customers does not automatically add 10,000 units of support capacity. Voice AI agents do.

The enterprise contract wins at ElevenLabs signal something specific about the maturity of the technology. Deutsche Telekom doesn’t run experiments at scale. When a company like that deploys voice AI for real customer interactions, it means the technology is production-ready — not just technically, but in terms of reliability, compliance, and output quality.

What This Means for Business

The $500M ARR milestone is not really a story about ElevenLabs. It’s a story about where voice AI sits in the enterprise stack right now.

Twelve months ago, deploying a voice AI agent that could handle real customer interactions felt like an enterprise-only capability requiring months of custom development. That’s changed. The platform layer is now mature enough that businesses of all sizes can deploy voice agents that sound natural, integrate with existing systems, and actually resolve customer requests rather than just routing calls.

For business owners thinking about customer service, sales support, internal help desks, or any workflow that currently requires humans to be available by phone or chat: the question is no longer whether voice AI is ready. The revenue numbers and the enterprise contracts confirm that it is.

The businesses that move in the next 12 months will have a structural advantage over those that wait. Not because voice AI is a secret — it clearly isn’t — but because the companies that deploy now will have 12 more months of data, process tuning, and refinement by the time their competitors get started.

Voice AI is no longer something you evaluate. It’s something you either deploy or fall behind on.

---

If you’re building with Claude or Codex right now, grab the free Working With Claude field guide. Thirty-two pages on the full ecosystem, Claude Code in depth, and how to roll agents out properly. Get the free guide.