AI Investing Tools: Value Over Features, Says Laura Thomas

Every brokerage and budgeting app seems to be racing to bolt an AI badge onto its product. Chatbots, sentiment scanners, auto-generated portfolio summaries — the feature list keeps growing. But Laura Thomas of Longbridge recently made a point that cuts through the noise: AI in investing has to deliver real value to the person using it, not just add another tab to click.

It’s a simple idea that’s surprisingly rare in practice. A lot of what’s marketed as “AI-powered” investing is really a way to make an app look modern. The harder question — does this actually help someone make better decisions, save money, or avoid mistakes? — often goes unasked.

Why the “features first” trap is so common

Product teams like features because they’re easy to demo and easy to put in a press release. “Now with AI-generated stock insights” sounds impressive on a landing page. Whether those insights change what an investor does, or whether they’re accurate enough to trust, is a much messier thing to measure.

Thomas’s argument reflects a maturing view of the technology. The novelty of talking to a chatbot about your portfolio wears off quickly. What sticks is whether the tool genuinely reduces friction or improves outcomes over time.

Features that look useful but often aren’t

• Generic AI summaries that restate a company’s earnings report without any interpretation you couldn’t get from the headline.
• Sentiment scores pulled from social media that have little proven link to future returns.
• Chatbots that answer questions confidently but can’t cite where the answer came from — or get it wrong.
• “Personalized” recommendations that are really the same handful of popular stocks shown to everyone.

None of these are inherently bad. The problem is when they’re the whole pitch, and the value to the investor is assumed rather than demonstrated.

What real investor value actually looks like

Value shows up in outcomes and behavior, not in the technology itself. A useful AI feature usually does one of a few concrete things: it saves you time on work you’d otherwise do manually, it catches something you’d have missed, or it helps you act more consistently.

Some examples that pass that test:

• Fee and cost analysis: AI that scans your holdings and flags high expense ratios or overlapping funds gives you a specific, actionable saving.
• Tax-aware suggestions: Tools that identify tax-loss harvesting opportunities or the tax impact of a sale help with a decision that has real dollar consequences.
• Behavioral guardrails: Nudges that flag when you’re about to make an emotional trade during a market drop can protect returns more than any hot stock tip.
• Research synthesis with sources: An assistant that summarizes filings and links back to the exact document lets you verify instead of trust blindly.

The common thread is that each one connects to a decision the investor was going to make anyway, and makes that decision better or faster.

How to judge an AI investing tool for yourself

You don’t need to be technical to separate substance from marketing. A handful of questions will tell you most of what you need to know.

Ask what problem it solves

If you can’t say in one sentence what the tool helps you do differently, that’s a warning sign. “It gives me insights” isn’t a use case. “It tells me which of my funds duplicate each other” is.

Check whether it shows its work

Trustworthy AI tools cite sources, explain their reasoning, and make it easy to verify. Anything that hands you a conclusion with no supporting detail should be treated with caution — AI models still make confident mistakes, and money is a bad place to find out.

Look at what it costs you

Some AI features are free because your data is the product, or because they nudge you toward trades that generate fees. Ask how the provider makes money and whether the tool’s advice could be shaped by that.

Test it against something you already know

Run the tool on a decision you’ve already thought through carefully. If its analysis matches your reasoning or improves it, that’s a good sign. If it’s vague or wrong on familiar ground, don’t trust it on unfamiliar ground.

The bigger shift Thomas is pointing to

The first wave of AI in finance was about proving the technology could be added at all. The next wave — the one Thomas is describing — is about proving it deserves a place in how people actually invest. That’s a higher bar, and it favors companies willing to measure whether their features change outcomes rather than just engagement metrics.

For everyday investors, the takeaway is practical. Don’t be impressed by the AI label on its own. Judge these tools the same way you’d judge a human advisor: Does it help me understand my money, act more sensibly, and keep more of what I earn? If a feature can’t answer that, it’s decoration.

AI genuinely can help with investing — automating tedious analysis, spotting costs, and keeping emotions in check. But the value comes from the job it does for you, not the technology behind it. As the tools multiply, that distinction is the one worth holding onto.