According to PYMNTS.com, JPMorgan Chase has decided to move away from traditional proxy advisory firms, as reported on Wednesday, January 7. The bank will now rely on an internally built AI system named Proxy IQ for its shareholder voting analysis. This tool is designed to aggregate and analyze proxy data from roughly 3,000 annual company meetings. The immediate effect is the disintermediation of those external advisers, reducing cost and latency. The move signals AI’s deeper integration into the operational core of financial services, a trend PYMNTS tracks as “cognitive banking.”
Why this is a big deal
Look, proxy voting sounds boring. But for a giant like JPMorgan‘s asset management arm, it’s a huge, manual, and expensive process. They used to pay outside firms to read thousands of dense proxy statements and then tell them how to vote on things like board elections. Now, they’re basically saying, “We’ve got an AI that can do that faster and cheaper in-house.” Here’s the thing: it’s not just about saving a buck. It’s about control and speed. By cutting out the middleman, JPMorgan shortens the loop between getting corporate data and making a decision. In finance, speed is everything.
The broader trend: cognitive banking
This isn’t happening in a vacuum. JPMorgan’s proxy move is a textbook example of what’s being called “cognitive banking.” The idea is that AI, sitting on top of all a bank’s internal data, stops just answering questions and starts proactively suggesting actions. We’ve already seen this in fraud detection, where systems block transactions in real-time instead of reviewing them later. And in customer service, where chatbots are moving past simple FAQs to actually analyzing a customer’s situation. JPMorgan is just applying that same logic to a new, highly manual domain—governance and asset management. It’s a beachhead, and you can bet other big banks are watching closely.
The trade-offs and oversight
But let’s not get carried away. AI doing the grunt work doesn’t mean AI is making the final, principled decisions. Proxy voting has serious fiduciary and reputational weight. The AI can scan and flag inconsistencies across 3,000 documents in minutes—a task that would take a human team weeks. But the bank’s human governance teams still have to set the voting policies and be accountable for the outcomes. The promise is that those teams can now spend less time on data gathering and reconciliation and more time on actual judgment. So is this about replacing humans? Not really. It’s about reshaping their role from data processors to decision-makers. The system still needs a strong human in the loop, especially when things get complex or controversial.
What it means for the industry
So what’s the signal here? For one, it shows a level of institutional trust in AI that simply didn’t exist five years ago. Banks are notoriously risk-averse. Letting an AI system guide votes on billions in assets is a massive vote of confidence in the technology’s reliability. It also highlights a push for operational sovereignty—bringing critical functions back in-house where they can be controlled and optimized. For the proxy advisory industry, this is a direct threat. If JPMorgan’s bet pays off, other major asset managers will surely follow. Basically, we’re watching AI move from the edges of banking (chatbots, fraud alerts) right into its analytical and governance heart. And that changes everything.
