Franklin Templeton's digital assets lead said this week that the rise of autonomous AI agents will push demand for blockchain protocols built to handle machine-to-machine micropayments. The statement, made during a virtual industry event on July 22, 2026, signals that one of the world's largest asset managers sees a concrete use case beyond human-to-human crypto transactions.
The AI agent thesis
The argument is straightforward: as AI agents start acting independently — buying compute, renting storage, paying for API calls — they'll need to settle tiny payments at high frequency. Traditional payment rails aren't built for that. Blockchain protocols that natively support micropayments could become the default settlement layer for these autonomous agents.
Franklin Templeton's digital assets lead didn't name specific protocols. But the logic points to networks with low fees, fast finality, and native token support for microtransactions. Think layer-1s or layer-2s designed for high throughput rather than general-purpose smart contracts.
The timing isn't accidental. AI agent frameworks have matured rapidly in 2026. Several projects now let agents hold wallets, sign transactions, and pay for services without human intervention. That creates a real — not theoretical — demand for settlement infrastructure.
Franklin Templeton is a $1.5 trillion asset manager. When its digital assets team talks about a trend, it's not just academic. The firm has been building out its crypto capabilities for years, including tokenized money market funds and blockchain-based share trading. This latest comment suggests they're looking at the next wave: infrastructure for the agent economy.
No timeline was given, and Franklin Templeton didn't announce any new products tied to the thesis. But the statement adds weight to a growing conversation in crypto circles about AI agents as a real demand driver. Expect more traditional finance players to start asking questions about which blockchains can actually handle machine-to-machine payments at scale.
