AI in Action: UBS Tackles Banking Bottlenecks

Aug 7
2 min read

In the world of banking, people often chase trends. Today, the trend is artificial intelligence and UBS is riding that wave. Rob Karofsky, the head of UBS’s wealth business in the U.S., says they’re using AI to improve productivity. About 60% of their AI efforts are focused on making processes more efficient.

What UBS Is Actually Doing

UBS isn’t building some futuristic robo-advisor or trying to replace humans with machines. They’re focused on three specific areas:

Speeding up client onboarding : That means getting new clients into the system faster — paperwork, legal checks, setting up accounts.
Improving compliance (like KYC checks)Banks need to know who their clients are to prevent fraud and follow regulations. AI helps gather and check this information faster.
Giving advisors helpful promptsAI tools analyze data and suggest topics or actions that advisors might want to bring up with their clients — sort of like smart reminders.

All of this helps UBS advisors work faster and more efficiently. That’s the goal.

What This Isn’t

Let’s be careful not to imagine too much. UBS isn’t turning its wealth managers into robots. They’re not handing over investment decisions to algorithms. They’re using AI to take care of the boring, repetitive, and administrative tasks — the stuff that eats up time.

That’s not disruptive innovation. That’s good housekeeping.

Why This Matters

Banks are huge machines with lots of moving parts — and many of those parts are outdated. UBS said they’ve had over 1,150 legacy systems running. That’s a lot of complexity and cost.

When you use AI and technology to simplify these systems, you can:

Lower costs
Reduce errors
Speed things up
Free up humans to do higher-value work

If UBS executes this right, they’ll be more agile, serve clients better, and likely make more money. That’s capitalism at work.

But There’s a Catch

Even though UBS isn’t replacing people with AI, there’s a long-term tension here:

Will the bank eventually need fewer advisors?
Will some clients get “lightweight” service, mostly from machines?
Will AI start influencing investment advice too much?

These aren’t accusations — they’re just possibilities to watch for.

The point is: AI is a tool, not a solution in itself. The bank still has to make good decisions, treat clients well, and not let the technology make mistakes that humans wouldn’t.