Carl Slabicki

speaker

440 appearances 2 recordings 1 series first heard Sep 2025 last heard 12 Feb

Carl Slabicki’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months

Recordings per month over the last 12 months — 1 in all, peaking in Feb 2026 with 1.

Appearances

newest first · ▶ plays the moment

Off the Rails from the U.S. Faster Payments Council - FPC · 25 Sept. 2025 -- Carl Slabicki from BNY on cross border, AI, and open banking · 25 Sep 2025 podcast

So Eliza was actually the wife of our founder, Alexander Hamilton, 241 years ago.

So we named our platform over Eliza.

And for the last two years, we've been really focused on a couple of things.

So one has been kind of a cultural and education learning journey for BNY.

So all of the 50,000 people across the bank in all roles, not just engineering, but

operations, client service, relationship management, product management, et cetera, have all taken AI courses internally, learned about it, have access to ELIZA, and can actually build their own agents on this platform.

So we actually have hundreds of different initiatives

and agents across the bank from all different functions where people can go in and you don't have to be an engineer to do this, right?

I can go in and create an agent to actually do a function that I think is applicable to my job, whether that's product management service, something else.

And we've had this very kind of grassroots effort for people to really experiment, try things out, build their own agents, use the workbench, leverage it in kind of different aspects.

And we've seen a tremendous amount of success just in the last two years.

So if you think about our payments business as an example, we're already using it both in the operations capacity and the engineering capacity and in the front office capacity.

I'll give you a couple of examples.

So in payment operations, right, we have payments that go into sanctions review, right, to, you know, potential something on a payment kind of matches something in a sanctions list that's got to kick out for someone to review that.

Or it goes into a repair where, you know, someone needs to be able to say, hey, there's a missing city, country code, whatever it may be.

We actually have built digital employees that are AI agents in our operations team.

So they have an avatar, they have an email, they send them an email, they'll write you back, they have a manager, they have their company ID.

And we built them to really focus on the manual repetitive tasks that the people would be doing when these things kicked out from their straight-through process.

So these avatars are actually using kind of the rationale and the logic.

They'll be able to quickly, within seconds, pull data on history, web searches, these type of things, and make a highly confident recommendation to a human operator to process the payment much faster with better confidence.

Showing 281–300 of 440 · page 15 of 22