Or maybe we should title this post “Do Not Accept Payment” instead.
So, a long time ago I wrote about some of the very good work that was being done by the folks associated with the “chat bot” app DoNotPay. If you do not remember any of that, you can refresh through this link.
The latest developments in the world of DoNotPay provide more proof that you either die a hero or live long enough to become the villain. Joshua Browder appears to be the tech-bro working on his villain resume.
So far, most of this has played out on Twitter, so here’s a link to the first indication of Mr. Browder losing the plot: Joshua Browder on Twitter: “DoNotPay will pay any lawyer or person $1,000,000 with an upcoming case in front of the United States Supreme Court to wear AirPods and let our robot lawyer argue the case by repeating exactly what it says. (1/2)” / Twitter
And then here’s the second such thread where it escalates. (Mr. Browder has now apparently deleted the tweet, when I can find a screenshot, I will paste it below. But he posted yesterday that lawyers were complaining that a million was too low and upped the offer to $5 million.)
It probably goes without saying to anyone who regularly reads things here, but this is not an offer any lawyer could ethically accept.
While I have interacted with Mr. Browder on Twitter directly, but perhaps far too flippantly, let’s be serious about discussing a few of the ethical prohibitions that would come into play here.
My favorite to begin would be the violations of RPC 1.8 and RPC 5.4 in any jurisdiction tracking the ABA Model Rules. The key reasons those rules would prohibit what DoNotPay is proposing are because of the two things that are key to the deal — payment of money and a requirement that the lawyer say whatever the AI tell them to say.
Model Rule 1.8(f) prohibits a lawyer from “accept[ing] compensation or direction in connection with the representation of a client from one other than the client unless … (2) there is no interference with the lawyer’s independence of professional judgment or with the client-lawyer relationship.”
Model Rule 5.4(c) similarly demands that a lawyer “shall not permit a person who … pays the lawyer to render legal services for another to direct or regulate the lawyer’s professional judgment in rendering such legal services.”
There is no set of circumstances where a lawyer could agree to the terms Mr. Browder is proposing without inherently violating both of these provisions as ceding any ability to exercise professional judgment about whether to repeat what the disembodied voice says is itself the bargain being struck.
Other ethics rules that would be violated flow from the United States Supreme Court’s own rules, but the violations of 1.8 and 5.4 is insurmountable and would apply if this was undertaken in any court.
Based on the understanding that the United States Supreme Court does not allow lawyers to bring in certain electronic devices at all, doing what Mr. Browder proposes would also cause a lawyer to violate any rule patterned on Model Rule 3.4(c) which creates problems for lawyers “who knowingly disobey an obligation under the rules of a tribunal.”
There are also lots of rules governing the duties a lawyer owes to a client that might be hard to reconcile with accepting this sort of proposal, but some might try to argue that a client could give informed consent. Maybe. But the personal interest conflict created for the lawyer — the only reason that a lawyer would ever succumb to the offer … a check for $5 million in exchange for doing something obviously not in the client’s best interest — appears to be pretty clearly an unwaivable conflict under any version of Model Rule 1.7(b).
And, of course, since it seems pretty clear that the concept does not involve letting the Justices know what is taking place — because if you did they wouldn’t let it happen — the proposal also would require a lawyer to violate any version of Model Rule 8.4(c) since the lawyer would be “engag[ing] in conduct involving dishonesty … [and] deceit.”