The company that powers your two-factor auth codes just posted record profits, and AI is barely showing up in the numbers yet.
The Summary
- Twilio beat earnings expectations with record profitability and free cash flow, sending shares up sharply
- CEO Khozema Shipchandler says AI is currently "only a modest contributor to revenue" but sees agentic AI as a major long-term growth driver
- The company's bet: its global communications infrastructure becomes essential plumbing when AI agents need to talk to humans at scale
The Signal
Twilio's earnings beat matters because it's one of the clearest signals yet that the infrastructure layer for agentic AI is being built right now, quietly, inside companies most people think of as legacy cloud utilities. Twilio runs the pipes for SMS, voice, email, and video communications for hundreds of thousands of businesses. When an AI agent needs to send you a text, make a phone call, or fire off an email, it's probably using Twilio's APIs.
The interesting part is what Shipchandler didn't say. AI is a "modest contributor" today, which means Twilio's record profitability is coming from traditional business communications. The company is printing cash from boring infrastructure work while positioning itself as the connective tissue for the agent economy.
"Agentic AI is a long-term tailwind, not today's revenue driver."
Here's why that matters:
- Twilio already has global telecom relationships and regulatory clearances that take years to build
- When agents start handling customer service, appointment scheduling, and outbound sales at scale, they'll need reliable ways to reach humans
- Building that from scratch is expensive and slow, which means Twilio's moat widens as agentic AI adoption grows
The company's infrastructure advantage is real. Connecting an AI agent to the global phone network or SMS systems isn't trivial. You need carrier agreements in dozens of countries, compliance with telecom regulations, fallback systems when networks fail, and the ability to handle billions of messages without dropping packets. Twilio has spent 15 years building that. A startup trying to compete in the agent-to-human communication layer would be starting from zero.
What Shipchandler is signaling is that Twilio sees two revenue waves coming. Wave one is already here: businesses using AI to automate internal workflows still need to communicate with customers through traditional channels. Wave two is further out: fully autonomous agents that operate businesses end to end, where the communication layer becomes as critical as compute or storage.
The Implication
If you're building AI agents, pay attention to infrastructure companies like Twilio that are already profitable and sitting on network effects. The picks-and-shovels thesis still works, but the interesting plays aren't the ones hyping AI revenue today. They're the ones admitting AI is modest now while quietly preparing for the agent economy's communication needs at scale. When agents start making a million phone calls a day on your behalf, someone has to route those calls. Twilio is betting it's them.