When AI Gets an Email: What Digital Workers in Banking Mean for Courts and Law Firms
A recent Wall Street Journal article reported something that would have sounded outlandish just a few years ago. Apparently, some major banks are now giving artificial intelligence agents their own email addresses and Microsoft Teams logins. And these "digital workers" are not just assisting behind the scenes. They're being treated as members of the team. They perform operational tasks, communicate autonomously, and interact with coworkers much like any human employee would.
For those of us in the legal system, this isn't just interesting. It's deeply instructive.
Law tends to trail fintech when it comes to innovation. Electronic signatures, virtual hearings, even basic automation. Financial institutions typically lead the way, and the legal profession adapts a few steps behind. So when we see digital agents operating with full credentials and communication privileges in one of the most tightly regulated sectors of the economy, it's not just a signal. It's a preview.
The shift isn't about using smarter tools. It's about redefining what a "worker" is. When a bank creates a digital entity that logs in, sends messages, and operates semi-independently, we're not just talking about automation. We're talking about role replacement. Software that doesn't just assist with tasks but fills a seat on the org chart. And if finance can do that with its strict compliance culture, what happens when the legal system starts following that path?
The implications for law firms are immediate. We've already seen AI draft contracts, summarize depositions, and sift through mountains of discovery. But what happens when that AI has an email address? Or when it joins the firm's Slack channel and becomes a working member of the team?
That's where things get complicated. If a digital worker sends an email with incorrect advice or shares confidential information, who is responsible? If it communicates with a judge or court staff, is that an ethical breach? And if the AI gets something right, really right, does the supervising attorney still get the credit?
Now imagine how this could be implemented in courts.
Clerks' offices are already overwhelmed in many jurisdictions so the idea of assigning a digital assistant to handle basic intake, file routing, or scheduling sounds appealing. But once that assistant starts receiving and sending messages, once it becomes a "participant" in the process, we're no longer just modernizing. We're rethinking the architecture of the justice system.
Will courts give digital workers their own credentials? Will they appear in docket notes, case tracking systems, or internal memos? What happens when a digital assistant in a judge's chambers starts pinging an AI working at a law firm? Whose voice is it? Whose action is it? And what part of the record does it become?
We are not ready for this. And yet it's already happening in the banking industry.
According to the article, the bank reported on has moved fast but carefully, assigning each AI agent to a specific team and limiting its access. The goal is clear: avoid overreach while still capturing the productivity gains. But even in that narrow model, the AI is no longer a tool. It's a teammate. It has a name, a login, a digital presence. And potentially, operational influence.
The legal system must prepare for this shift with the seriousness it demands if we go down this road. Because once AI gets its own email and starts showing up in your inbox, it's not just a tool anymore. It's a colleague. And that changes everything.
Subscribe to my Substack newsletter today so you don’t miss out on a post. https://judgeschlegel.substack.com
