Will an AI Takeover Help Democratize Finance?

There is a lot of fear right now about what AI is doing to human work. I don't know anyone who feels entirely immune. We don't know what our world will look like in less than 5 years, and it's rare that we feel that level of uncertainty at such a large scale. But I want to tell you about a piece of research that made me genuinely hopeful, because it highlights something I've been paying a lot of attention to: AI is already disrupting finance, especially portfolio management- and the people who stand to benefit most may be the investors who have historically been underrepresented.

There's nothing new about using technology to more effectively manage portfolios. Over the last decade, we've seen how index funds have taken over mutual funds as a preference for retail investors. What I'm talking about is the difference between actively and passively managed funds. Let's do a quick refresher-

Mutual Funds vs. Index Funds: A Quick Primer

For decades, the only way everyday people invested in the stock market was through mutual funds. These are professionally managed portfolios where a human expert (or team of experts) actively decides which stocks to buy and sell on your (the investor's) behalf. For a while, mutual funds were revolutionary because they gave everyday investors the chance to invest in funds that were made up of various company stocks- and they were hand-picked by experts. It was way better than investors picking individual stocks themselves. The catch was: wrapped into the cost of these funds were the fees charged by these professional managers. Some fees soared in the 10-15% range, but dropped pretty quickly as more competition came around, and then dropped even more when index funds entered the equation.

Index funds offer investors the benefit of diversification through investing in funds (rather than individual stocks), but rather than the funds having human managers, they use technology to track an index in order to decide which stocks should go in the fund. One of the most common examples is VOO- an index fund ​we've talked about ​here before. It's the Vanguard index fund that tracks the S&P 500, which is an index that trades the 500 biggest US companies. An index fund tracking the S&P 500 would automatically buy a slice of every company in it. No stock-picking and no guesswork, which means dramatically lower fees (often under 0.1%).

Over time, when we had enough data to track the results of index funds vs. mutual funds, most actively managed mutual funds failed to outperform their index fund equivalents, after fees, in long time horizons.

Because of this, over the past two decades, money has flooded out of active mutual funds and into index funds. By 2024, index funds held more U.S. stock market assets than actively managed funds for the first time ever. This was a pretty seismic shift driven largely by retail investors voting with their wallets.

I'd argue that the the shift from mutual funds to index funds brought more than just fee-savings for investors; I think this shift is what played a huge role in the personal finance information boom, and the ​FIRE​ movement. Suddenly, everyday investors had a direct, transparent view of their own portfolios. There was no black box, no expert speaking in jargon behind closed doors. You could see exactly what you owned, watch it move with the market, and actually understand why (I admit, by the way, that the internet provided a lot of this transparency for both mutual funds and index funds- but that's just another way that technology changed the game).

This transparency gave way to communities. On blogs, podcasts, Reddit threads, and YouTube channels, people shared strategies, asked questions, and realized for the first time that building wealth in the market was something they could do themselves, on their own terms. The "leave it to the experts" era began to crack open, and in its place emerged a generation of retail investors who were engaged, curious, and less willing to be talked down to.

It's worth pausing on who that mattered most for. People who previously hadn't been given permission to be in that "block box" room of experts: People of Color, people who didn't come from money, and Women- the folks who have historically been underrepresented in investing spaces and more likely to be steered toward conservative or passive financial roles. We suddenly had access to (much of) the same information, tools, and markets as (most) everyone else. The playing field still hasn't become perfectly level- but it tilted!

The reason this all matters in the context of AI is because recent research suggests that the active management industry's problems are about to get even worse.

A ​new paper​ published by researchers at Harvard, Wharton, and DePaul University explores how the researchers trained an AI model on the trading histories of over 1,700 active mutual fund managers. The model learned every buy, hold, and sell decision across more than three decades. After this, the AI correctly predicted 71% of their trades, without the manager making a single decision. For some managers, it was nearly all of their trades.

In other words: a significant portion of what highly paid fund managers do is… habit. Predictable, replicable, pattern-driven behavior that a machine can learn and mimic. What I love most about this is the result that the managers whose trades were easiest to predict actually underperformed their peers. The ones whose decisions were hardest to mimic were the ones that outperformed. Novelty and genuine judgment, it turns out, are what generate real returns. Novelty and genuine judgement, it turns out, are also very difficult for AI to replace (so far).

The Part I'm Most Excited About

If you've been here for a minute, you know that I believe investing can be one of the most powerful tools a woman can have- not just for building wealth, but for shaping the world she wants to live in, and that she wants to leave for her great-grandchildren. So, when I read this research, I kept thinking: if AI handles the rote, pattern-driven parts of portfolio management, what's left over is the human part. The values-driven part. The part that asks: how can the stocks in this fund best help the people and the causes we're meant to uphold?

I'm hopeful that AI can open up real possibilities here. When AI can accurately make the predictions humans have been making for decades, maybe it will ripple into the good part of investing:

✦ Identifying companies that have promise to make some of the biggest impact in climate, labor, and equity, and redirecting capital to them.

✦ Elevating the humans who are already contributing the nuanced, ethical, and judgement-based feedback to funds and companies.

✦ Allowing funds that invest in values-based initiatives to be more successful.

✦ Making investing more accessible to people who've historically been locked out.

✦ Shifting power in finance away from a historically homogenous industry, and toward a more diverse, values-led model.

Finance has long been dominated by a narrow demographic. If the mechanical, pattern-driven work gets automated, and what remains is the emotional intelligence, relational wisdom, and values alignment that guides truly good financial decisions, then I'll take that as a huge win. That, my friends, is terrain where women have always excelled, and have often been underestimated. I believe this shift creates real runway for gender and racial equity in how wealth is built and managed. Tools that strip away expensive habit and surface might also be able to capitalize on values, and genuine insight. Maybe we're about to witness investing get smarter, fairer, and more human, not less.