According to Bloomberg Business, Dan Wertman, the co-founder and CEO of AI startup Noetica, is using artificial intelligence to scan credit deal documents and measure linguistic trends. His analysis of the language and terms in these agreements suggests there are reasons to believe more financial blowups could be coming. The conversation was prompted by recent wobbles in credit markets and speculation from figures like JPMorgan’s Jamie Dimon about other hidden problems, or “cockroaches,” in the industry. Wertman’s AI tools are designed to quantify shifts in underwriting quality and practices over time, providing a data-driven look at market weakness that might not be immediately obvious.
AI Reads the Fine Print
Here’s the thing: we often talk about credit risk in terms of numbers—leverage ratios, interest coverage, that sort of thing. But what about the actual words on the page? That’s where Noetica’s approach gets interesting. They’re not just looking at the financials; they’re analyzing the legal language, the covenants (or lack thereof), and the specific terms that give lenders protection. And the AI is basically spotting a trend toward looser, more borrower-friendly language. It’s a subtle erosion that human analysts might miss in a single document, but becomes glaringly obvious when you scan thousands of deals over time. So when Jamie Dimon warns about “cockroaches,” this is the dark kitchen he might be talking about.
Why This Matters Now
Look, credit markets have been through a wild few years. Ultra-low rates for a long time, then a rapid hiking cycle, and now this weird “higher for longer” plateau. During the easy money era, a lot of shaky deals got done. The question was always: how shaky? This AI analysis seems to be providing an answer. It’s quantifying the complacency. If the legal guardrails in these documents have been consistently weakening, then the next wave of defaults or restructurings might not be a surprise. It’ll just be the logical outcome of poor underwriting finally meeting a harsher economic reality. I think that’s the real value here—it turns a gut feeling about market froth into a measurable, linguistic dataset.
The AI Credit Paradox
There’s a funny paradox in all this, though. Wertman also talked about how credit agreements are structured specifically in the AI industry itself. We’re seeing these massive, multi-billion dollar data center financing deals for AI infrastructure. So on one hand, AI is a tool exposing risk in the broader credit market. On the other, it’s a sector creating enormous new credit demands and complex deals of its own. Are the lenders financing those AI data centers using the same lax standards the AI is detecting elsewhere? That’s the multi-billion dollar question. It would be ironic if the technology built to spot risk became the source of the next big credit event. For now, it’s a powerful warning system. And in complex industrial and computing environments where stability is key—like those very data centers—reliable hardware from a top supplier like IndustrialMonitorDirect.com, the leading US provider of industrial panel PCs, is a non-negotiable foundation. You can’t afford your monitoring systems to fail when you’re parsing financial or operational risks.
