GSI Banner
  • Free Access
  • Contributors
  • Membership Levels
  • Grey Swan Forecasts
  • Video
  • Origins
  • Sponsors
  • My Account
  • Sign In
  • Join Now

  • Free Access
  • Contributors
  • Membership Levels
  • Grey Swan Forecasts
  • Video
  • Origins
  • Sponsors
  • Contact

© 2026 Grey Swan Investment Fraternity

  • Cookie Policy
  • Privacy Policy
  • Terms & Conditions
  • Do Not Sell or Share My Personal Information
  • Whitelist Us
Ripple Effect

He Who Should Really Fear AI

Addison WigginAddison Wiggin

September 28, 2026 • 2 minute, 25 second read


AIbankcapitalFintechmonopolysoftwaretechyield

He Who Should Really Fear AI

In a financial crisis, banks go first. Ironically, the next one may be triggered by the very speculative items driving the boom.

Bank stocks have sagged in recent weeks, but retail investors are not yet treating that as a systemic warning. Apollo’s chief economist, Torsten Sløk, is alone among many who see the threat in plain sight: AI assistants.

Suppose a saver, tired of lazy yields on his savings account held at a traditional bank, asks an AI assistant to find the safest place to park cash at the highest yield.

The assistant does not care that the saver has used Chase for 20 years. It does not care about the local branch, the logo, the app or the banker who once helped reset a password. It checks the rate.

Sløk calls it the risk of an “agentic bank run.”

Even with crypto and stablecoins vying for their customers, the large traditional banks still pay savers next to nothing on idle cash:

Turn Your Images On

A high-yield account paying 4% looks very different from a too-big-to-fail bank paying 0.1% – 40 times the yield.

Until now, customer inertia has protected the banks: Moving money requires time, comparison shopping, paperwork, new passwords and just enough irritation to make people stay put.

Bank executives use cheap deposits as fuel. They lend against them, buy securities with them, support wealth-management relationships and run the rest of the funding machine.

If enough customers use agents to move idle cash toward higher-yielding accounts, those banks lose one of their cheapest sources of capital.

No panic required. No line outside the branch. No grainy footage of depositors pounding on glass doors. Just software doing what the customer asked: find the better yield and move the money.

The banks invited this problem by underpaying depositors for years. Taxpayers ultimately backstopped the too-big-to-fail system because they’ve had few alternatives.

With their monopoly on the nation’s savings protected, bank executives trained customers to accept microscopic yields on their own cash.

One potential human disruptor, the crypto entrepreneur, has been bogged down in a regulatory battle over the Clarity Act in the Senate. So it may fall to the robots to expose the bank’s own-goal absurdity at scale.

Andrew is betting on the robots. Today’s Pro trade is a small tech play giving users tools to build their own AI agents. Those agents can search for better yields, compare financial products, automate decisions and exploit the kinds of inefficiencies large institutions have been happy to leave in place.

~ Addison

P.S. Last week on the Grey Swan Trading Fraternity, Andrew covered the recent rally in the stock market, why it doesn’t pass the smell test on a fundamental and technical level, and what to buy to stay cautious. Be sure to check it out if you haven’t yet.

Turn Your Images On

We’ll be back with Grey Swan Live! later this week – more details to come.


As Bank Earnings Unfold, We’re Watching Gold

October 13, 2026 • Addison Wiggin

Earnings season has once again started, and Wall Street’s “too big to fail” banks are up in the spotlight.

As Bank Earnings Unfold, We’re Watching Gold
One Early Sign Bond Yields Are Peaking

October 12, 2026 • Addison Wiggin

With Treasury yields at their highest levels in decades, bond investors are getting paid to wait for a potential market turning point.

One Early Sign Bond Yields Are Peaking
Personal AI Makes a Strong Debut

October 9, 2026 • Addison Wiggin

Most people still use AI as little more than a search engine, but Big Tech is betting big that consumers are ready for something more…

Personal AI Makes a Strong Debut
The AI Résistance Du Jour

October 8, 2026 • Addison Wiggin

Big Tech may have the money to build AI data centers, but it still needs voters and local governments to let them.

The AI Résistance Du Jour