JPMorgan to Deploy Hour-Long Autonomous AI Agents in 2026

JPMorgan Chase will deploy artificial intelligence agents capable of operating autonomously for extended periods later this year, representing a significant advance in how corporations use AI technology, according to details shared exclusively with CNBC.

The bank's planned rollout centers on what Derek Waldron, JPMorgan's chief analytics officer, describes as "long-running autonomous agents" — systems that can work independently for an hour or two rather than just a few minutes. These agents will manage workflows spanning multiple steps and different software platforms, functioning more like digital workers than simple task-completion tools.

Why it matters

This deployment signals that AI agents are clearing the security and governance barriers that have kept them out of large financial institutions. With JPMorgan's nearly $20 billion annual technology budget and position as the largest U.S. bank by assets, its adoption could accelerate enterprise use of autonomous AI systems across industries. The shift from cost-cutting to revenue generation also marks a strategic inflection point in how corporations view AI investment.

From task automation to workflow management

Waldron told CNBC that recent improvements in AI reasoning capabilities have transformed these systems from individual workers into team managers. They can now break down complex problems, delegate activities, and maintain what he calls "intellectual coherence" over longer timeframes.

Several technical advances enable this expanded capability: AI models can now write code, control web browsers, and interact directly with desktop software. While examples like Anthropic's Claude Code and OpenClaw have demonstrated long-running agents publicly, JPMorgan's implementation suggests the technology is ready for enterprise deployment despite lingering security concerns.

Waldron projects that agents will eventually maintain coherence for "multiple hours, then days, then weeks."

Revenue impact beyond cost reduction

The bank has already seen measurable results from AI deployment in revenue-generating roles. In private banking, AI systems analyze market activity, client positions, and research overnight, allowing bankers to concentrate on client relationships. JPMorgan reports a 20% increase in gross sales attributed to these tools and believes they could expand individual banker client coverage by up to 50%.

While CEO Jamie Dimon has acknowledged that some workers will be displaced by AI — with the bank preparing training and redeployment programs — Waldron emphasized that companies increasingly recognize AI's revenue potential beyond cost cutting. "For enterprises to win with AI, it's not about cutting the maximum number of jobs," he said. "It's all about trying to create a sustainable competitive advantage."

Implications for software vendors

The technology shift is also changing JPMorgan's build-versus-buy calculus for software. The bank now evaluates more carefully whether it can develop capabilities internally rather than purchasing from external vendors.

"The moat around certain types of software companies is most certainly diminished versus where it was in the past," Waldron said, suggesting potential pressure on traditional enterprise software providers.

These details were first reported by CNBC.