The Update: Retail Investors Are Now Trusting AI Tools With Millions

Our previous coverage tracked retail investors piling into Nasdaq calls at the fastest pace since 2021. That was about momentum and risk appetite. This is something different.

Retail investors aren't just buying AI stocks. They're trusting AI tools with their actual money. Millions of dollars of it. And the tools themselves are warning users not to.

The Numbers

Boaz Sobrado reported in Forbes on October 21, 2025 that a weekend side project called Rallies.AI — built by two software engineers, Hira and her husband, in their spare time — now monitors $40 million in retail investment portfolios. One single user connected an $8 million portfolio to the platform. The tool has processed nearly 50,000 user questions in just months.

Meanwhile, Robinhood launched Cortex, a new AI investment assistant announced at their recent Gold keynote. The company explicitly warns the tool "may not be accurate." User reaction: immediate excitement anyway.

Trading 212 rolled out an AI portfolio analysis feature. Users on their own forums call the analysis "wildly inaccurate." Usage remains high.

The pattern is consistent: the disclaimers don't matter. People want the validation.

Why This Is Happening

Sobrado's Forbes piece nails the psychology. Retail investors have spent decades being told they can't beat professionals. They're not irrational — they're isolated. They've been posting portfolio screenshots on Reddit and begging strangers for advice. AI fills that void with something that feels more authoritative, even when it isn't.

This isn't stupidity. It's what happens when financial advice costs $300 an hour and most people can't afford it.

Feeling less alone and making good decisions are different things.

Regulators Just Flagged This — Specifically

The Ontario Securities Commission, working with the Behavioural Insights Team, published a formal research report on AI and retail investing. According to the OSC, their study identified three core AI use cases hitting retail investors right now: decision support, portfolio automation, and scams/fraud.

The benefits they acknowledge are real. AI can cut the cost of personalized advice. It can reach investors who have zero access to traditional financial guidance. Done right, it can improve diversification and risk management.

But the OSC also flags serious risks — model bias baked in by developers, potential for automation to remove human judgment at exactly the wrong moment, and an entire fraud category where bad actors are exploiting the AI "buzz" to run scams on retail investors who assume AI-branded tools are legitimate.

The OSC's behavioral experiment tested something specific: does the source of an investment suggestion — AI, human, or a blend — change how much an investor follows it? The results aren't fully published in what's available, but the fact that they ran this experiment signals what they suspected.

The Governance Angle

Most mainstream financial outlets are covering AI investing as a market story — what stocks to buy, what the rally means, whether this is 1999 again. The shareholder governance dimension has received far less attention.

The Shareholder Association for Research and Education (SHARE) just released a report called Investor Advocacy on Artificial Intelligence, according to Investment Executive. Juana Lee, SHARE's associate director of corporate engagement, stated plainly: "There are no guardrails in place to ensure that the development and deployment of these AI systems are responsible and ethical."

SHARE isn't just writing white papers. They're engaging executive teams directly and pushing retail shareholders to use their votes to force accountability.

Their materiality framework covers financial risk, yes — but also bias from bad training data, human rights exposure, environmental costs of AI infrastructure, and employment disruption. Lee argues these aren't soft ESG concerns. They're material financial risks that will hit portfolios.

The core point — that companies deploying AI face real legal and reputational liability from that deployment, and that investors are currently pricing in zero of that risk — is difficult to dispute.

What Coverage Is Missing

Financial media is treating this as two separate stories: the AI stock rally and AI regulation. They are not separate stories.

Retail investors are simultaneously betting on AI companies and using AI tools to make those bets. The same psychological vulnerability — wanting validation, wanting to feel less alone in complex markets — is driving both behaviors.

The OSC study and SHARE's governance push represent something the rally coverage ignores entirely: institutional awareness that this is getting out ahead of any protective framework.

Bloomberg's paywalled piece on retail investors and the AI dilemma confirms this tension exists. But paywalled analysis doesn't reach the Robinhood user who just connected their life savings to Cortex.

The Gap

If you're using AI tools to manage your investments, understand what you're actually doing. You're using experimental software, built by companies that have legally protected themselves with disclaimers, to make decisions with real money.

The tools might be helpful. The OSC research acknowledges that. But helpful and accurate are different things, and no regulator has signed off on any of these platforms as meeting an advice standard.

The guardrails don't exist yet. The money is already flowing.