October 3, 2026
Bonus Content: The COBOL Goldrush Is Minting New Winners, and Crushing IBM
Editor’s Note: Larry Benedict – the hedge fund legend who beat the S&P 500 by 18 times in 2025 and made his clients $95 million during the 2008 crisis – says Trump’s installation of a new Federal Reserve chair is triggering the most significant shift in U.S. markets in nearly 20 years. He has already identified the one ticker he believes will be at the center of the money flows – and he’s revealing it completely free. Read more below…
Dear Reader,
Grab a pen and write down this ticker: TLT.
It could be the single most valuable ticker you hear about all year.
Beginning May 2026, billions of dollars could pass through it.
But before you rush out and buy it… WAIT.
There is a very specific way you must play this ticker if you want to make real money from it.
Do it wrong, and you’ll only capture a fraction of what’s possible.
Do it right, and you could double your money in a matter of days.
I know, because I’ve done exactly that before.
My name is Larry Benedict, and I’ve been trading TLT for years.
In that time, I’ve watched a 4% move in this ticker turn into a 117% gain for my readers who followed my recommendation – in just a matter of days.
And it’s all because of the very specific way I trade it.
Discover how to access exactly how I trade this ticker – and why right now is the best setup I’ve seen in years – by watching this exclusive, free briefing.
Click here to learn how to access my complete TLT playbook.
Regards,
Larry Benedict
Founder, The Opportunistic Trader
P.S. The current setup on TLT is more attractive than I have seen in years – but it won’t last forever – so if you want to learn how to position for what could be some of your best gains of 2026, click here.
The COBOL Goldrush Is Minting New Winners, and Crushing IBM
Anthropic published a blog post in February about using Claude Code to modernize COBOL. Within hours, IBM’s stock dropped about 13%, its worst single-day loss since October 2000. The move erased roughly $30 billion in IBM market capitalization. The market’s instinct was directionally right. Its diagnosis was incomplete.
The real question is not whether AI agents can rewrite legacy banking and insurance code. They clearly can. AI-driven code refactoring has achieved 93% COBOL-to-Java conversion accuracy in a research prototype, and Microsoft has said more than 80% of Fortune 500 companies use active AI agents built with low-code/no-code tools. The question is who captures the fee stream that explodes out of that capability.
IBM’s position is more complicated than a 13% drop suggests. IBM’s senior vice president of software, Rob Thomas, argued publicly that “AI strengthens the mainframe case, it does not weaken it.” He has a point, partially. COBOL supports more than 40% of online banking systems, 80% of in-person credit card transactions, and 95% of ATM transactions, according to IBM itself. Nobody is pulling those cores overnight. But the consulting fees that IBM charged to understand, document, and slowly migrate that code are now being compressed into weeks by agentic platforms.
McKinsey documented a bank modernizing 20,000 lines of legacy code where a project originally estimated at 700 to 800 hours saw that estimate cut once an orchestrated set of AI agents took on the work, with the relationship-mapping step alone dropping from 30 to 40 hours down to about five. That is not a productivity gain for IBM’s consulting arm. That is a structural repricing of its labor model.
The winners look different from what the IBM trade implied. Blitzy, valued at $1.4 billion in May 2026, modernizes legacy enterprise codebases at scale, orchestrating thousands of AI agents working in parallel to understand, modify, and validate code across systems often built decades ago. A large bank applied an agentic AI approach to modernize its legacy core software, with different agents handling documentation, code generation, peer review, and integration testing, leading to more than a 50% reduction in development time and effort.
The failure risk is underappreciated. A growing menu of vendors is promising that agentic AI can solve mainframe modernization on its own. Gartner has warned that more than 70% of AI-driven mainframe exit projects started in 2026 will fail, driven largely by technology leaders overestimating what generative AI can actually do. That failure rate creates its own opportunity: firms that pair AI tooling with senior architectural oversight will charge a significant premium over pure-software plays.
Stocks to Watch
- IBM (IBM): IBM’s generative AI book of business has grown to more than $12.5 billion since inception, suggesting the company is reorienting faster than the stock decline implies. The consulting margin story is the risk to watch.
- Accenture (ACN): Growing spending on application modernization, cloud enablement, and cybersecurity-as-a-service represents a solid growth opportunity. Accenture’s scale gives it an advantage in the high-failure-rate environment where clients want accountability.
- Cognition (private, Devin): The company has said 89% of code committed by its engineers was committed by Devin, and TechCrunch reported a $26 billion valuation after a $1 billion raise in May 2026. The next test is whether that translates to regulated financial clients who cannot afford a quiet data error.
