In finance, AI owns the math, but humans own the gut

One of my Bentley University students put it plainly the other day: “AI taking entry-level jobs is a ‘when,’ not an ‘if.’ But in venture capital, 70% of the decision is reading the founder and team—and that’s something AI can’t do.”

That simple breakdown , 70% people, 30% product—flips the usual narrative about finance. For decades, finance was defined by numbers. Analysts lived and died by the spreadsheets. Today, AI can run discounted cash flows, parse a term sheet, and size a market faster than any junior associate. But if you talk to people in venture capital, they’ll tell you the math has never been the most important part. The numbers matter, of course, but the difference between betting on a future unicorn and losing it all is whether you can read the humans across the table.

AI’s takeover of the “crunching” jobs

Students see what’s coming. The entry-level finance roles that once trained armies of analysts are increasingly exposed to automation. AI models can scan thousands of comparable companies in seconds, build slide decks, even flag anomalies that a first-year hire would have spent a weekend catching. In fact, McKinsey estimates that nearly half of finance tasks could already be automated by existing AI tools. What was once seen as a rite of passage—the long hours bent over Excel—may soon look as outdated as typewriters in an accounting office.

That’s why many students I teach don’t just worry about if AI will change their jobs. They assume it already has. The conversation now is how to build careers in finance when machines are better, faster, and cheaper at the very tasks that used to get you in the door.

The venture capital exception

Venture capital, especially at the earliest stages, offers a counterintuitive lesson. The math can only take you so far. Market sizing, revenue projections, even technical due diligence—all of it is valuable, but none of it predicts success the way the founder does. Investors I’ve spoken with put it bluntly: you’re betting on people, not just products.

That’s the essence of the 70/30 rule my student repeated: 70% of the investment decision is about the team. Only 30% is about the idea. Surveys of venture investors back this up. First Round Capital, for example, has consistently found that founder quality outranks product or market sizing in early-stage decisions. You want founders with resilience, persuasion, and the grit to pivot when the idea inevitably changes. AI can tell you the addressable market. It cannot tell you if this particular founder has the charisma to convince skeptical investors, the judgment to hire the right people, or the sheer stubbornness to keep going after three failed prototypes.

As one early-stage investor told me recently, “AI can show me a founder’s track record in five seconds, but it can’t tell me whether they can read a room or take a punch. That’s still my job.”

Why Gen Z gets it

What strikes me is how quickly students are internalizing this. Rather than seeing their careers in finance as doomed, they are recalibrating. They want to be the people who can build relationships, persuade others, and trust their instincts. They’re preparing not to be number crunchers, but decision-makers.

In class discussions, they talk about internships where AI already plays a role in screening deals. Their takeaway isn’t despair—it’s clarity. They see that the irreplaceable part of the job is not working the models but reading the room. They know AI may one day suggest which startup should succeed, but it won’t sit across the table from a founder at 11 p.m., hear the quiver in their voice, and know whether that’s nerves or conviction.

Gen Z, often criticized for being “too soft,” may be better positioned than we think. They’ve grown up digital. They understand the strengths of machines, but they also see the limits. They’re comfortable letting AI do the heavy lifting if it means their human skills rise in value.

Lessons for the future of work

This shift should force us to rethink not only finance, but the future of work itself. If venture capital is a case study, the lesson is that industries once defined by quantitative rigor may end up placing even greater value on qualitative judgment. Harvard Business Review has made a similar point, noting that as AI scales technical analysis, “soft skills” are fast becoming the hardest skills to replace.

The irony is rich: the most number-driven fields may be the ones where numbers matter least. And if that’s true, Gen Z might just be the generation that restores humanity to the financial world—not by rejecting technology, but by mastering what machines can’t.

Back to the 70/30 rule

That brings me back to my student. Their comment wasn’t just about venture capital. It was about what comes next in finance, and perhaps beyond. If AI eats the numbers, the real work will be reading people.

AI owns the math. But the gut—the judgment, empathy, and intuition that turn data into decisions—still belongs to us.

In the end, the algorithms will get faster, but the best investors will always pause, look across the table, and trust what no machine can calculate: the human pulse of possibility.

View the full article