AI Investing Statistics (2026)

A 2026 survey of 938 US retail investors found that 62% now use AI tools to help with investing decisions, split between 23.6% who use them regularly, 27.4% occasionally, and 11.5% who have tried them. Another 21% are considering it, and only 16.6% rule it out.

Adoption runs higher among wealthier investors: HSBC's global study found 57% of affluent US investors, and 73% globally, use AI for financial tasks.

The generational split is stark. HSBC found 63% of US Gen Z and millennials actively use AI in investing, versus just 31% of Gen X and boomers.

Younger investors also trade far more actively, which pairs with heavier reliance on AI tools for fast information.

AI is mostly a research assistant, not an autopilot. Among AI users, 62.4% use it to research stocks, 35% to understand market news, 34.4% to generate trading ideas, and only 21.7% to make portfolio decisions (see the table below).

By tool, 53.5% use general chatbots like ChatGPT, 30.6% use dedicated research or screening tools, and just 4.5% use automated trading algorithms.

Optimism about AI as an investment theme falls sharply with age. The Motley Fool found 71% of Gen Z and 69% of millennials are bullish on AI stocks, versus 58% of Gen X and 52% of boomers (see the chart below).

The enthusiasm shows up in plans, too: 68% of Gen Z intend to increase their stock investments in 2026, versus 39% of boomers.

Trust is cautious. Only 3.8% of investors completely trust AI investment analysis and 19.8% mostly trust it; a majority, 53.5%, trust it somewhat but verify elsewhere, while 22.9% distrust it (see the chart below).

That caution is well-placed: AI models can hallucinate figures and cannot see your actual holdings unless you connect them.

Investors explore with AI but rarely let it decide. Only 7% of US investors cite AI as the single most influential factor in their last decision, and just 19% say AI was the source of their last investment idea, versus 59% who point to a financial professional.

Even so, 86% attribute at least some influence on their returns to AI, and on average credit it with about 35% of those returns.

Researching with AI is not the same as trading on it. Among investors, 26.6% have followed an AI trade idea multiple times, but 43.7% have never acted on an AI recommendation at all.

The direction of travel is clear, though: 37.8% expect to use AI much more for investing in the future.

Most investors want both. In HSBC's study, 38% of US investors prefer a hybrid AI-plus-advisor approach, rising to 50% of Gen Z and 44% of millennials.

Even heavy AI users lean on people: 77% still want reassurance from a human advisor and 68% seek human strategic expertise.

AI shifts confidence and risk appetite. About 48% of US investors say AI makes them feel more in control of their investments, versus 31% who feel less in control.

It also nudges risk-taking: 44% of US investors (49% globally) say AI makes them more willing to take calculated risks, which cuts both ways for outcomes.

Self-reported results are positive but modest. About 65% of AI users say it improved their performance (17% significantly, 48% somewhat), 33% saw no difference, and 1.9% said it made things worse.

The most-cited benefit is speed: 39.5% say faster market-data analysis is AI's main advantage, followed by fewer emotional decisions (14.6%) and spotting opportunities earlier (14%); 19.8% see no real advantage.

The worries are specific. Among investors, 38.9% worry about incorrect or misleading AI recommendations, 24.2% about market herding if everyone acts on the same signals, and 21% about over-reliance on automation; only 7.6% have no concerns.

These map onto real risk, which is why verifying AI output against a primary source remains essential.

Automated investing is now a mass market. US robo-advisors are projected to manage about $1.67 trillion in 2025, rising toward $1.91 trillion by 2029, and the number of robo users worldwide is forecast to reach roughly 34 million by 2028.

Globally, robo-advisor AUM is forecast near $11 trillion in 2026, and about 73% of wealth-management firms have deployed some form of AI robo-advisor.

The market is dominated by incumbents. Vanguard leads with about $300 billion in AUM, followed by Schwab Intelligent Portfolios ($89.5B), Betterment ($56.4B), and Wealthfront ($42.9B); the ten largest US platforms hold over $500 billion combined (see the chart and table below).

Acorns has by far the most clients, about 7.8 million, at a small average balance near $2,150, reflecting its spare-change model.

Bigger does not mean better returns. Among tracked platforms through late 2025, SoFi's robo posted the highest one-year return at about 14.1% (16.6% annualized over three years), ahead of Vanguard Digital Advisor's 10.1% and 14.3% (see the table below).

Returns depend heavily on each portfolio's stock-and-bond mix, so compare at the same risk level rather than by headline number.

Beyond retail, AI is embedded across financial services. By various industry estimates, about 85% of financial firms now use AI, 90% for fraud detection, 82% for algorithmic trading, and 64% of US banks for anti-money-laundering (see the table below).

Other common uses include knowledge management (49%), accounts-payable automation (37%), and anomaly detection (34%).

The most mature enterprise use of AI in finance is catching fraud. AI flags fraud an estimated 300 times faster than rule-based systems, at 90 to 99% accuracy versus 58 to 70%, cutting fraud-detection operating costs by around 60%.

It is sorely needed: more than 75% of US firms experienced payments fraud in 2025, and adoption of AI to fight it still lags the threat.

The money behind the trend is large and growing fast. The AI-in-fintech market was valued at about $36.6 billion in 2026, with forecasts near $99 billion by 2031 (see the table below).

The overall AI-in-finance market sits around $21 billion in 2026 by some measures, with estimates varying widely by how the category is defined.

Generative AI is the fastest-growing slice. The generative-AI-in-financial-services market is projected to grow from about $2.96 billion in 2025 to $25.7 billion by 2033, a 31% annual rate, and AI 'agents' in financial services are a new roughly $985 million market in 2026.

Adoption is still early: about 95% of generative-AI implementations in finance remain in pilot rather than at scale, partly due to regulatory caution around explainability.

Machines already run the market's plumbing. By industry estimates citing JPMorgan data, roughly 60 to 80% of US equity trading volume is algorithmic, and high-frequency trading alone accounts for an estimated 50 to 55%.

The concentration is extreme: about 2% of firms are estimated to generate the majority of high-frequency volume, and more than 90% of institutional trading desks use algorithmic execution.

AI has also supercharged fraud. The FBI's 2025 report logged a record $20.9 billion in cybercrime losses, including $8.65 billion in investment fraud and, in its first dedicated AI section, about $893 million tied to AI-enabled scams like deepfakes and voice clones (see the table below). Losses by Americans 60 and older reached $7.7 billion, up about 60%.

The FTC separately put 2024 total fraud losses at $12.5 billion, including $5.7 billion from investment scams, with a median loss of about $9,000 among the 79% of investment-scam victims who lost money, and warned that AI is making these schemes more convincing.

The data points to a clear posture: use AI to research, compare, and stress-test, but keep a human in the loop for decisions, exactly what most investors already do.

Two rules matter most. Verify any number an AI gives you against a real source, because models hallucinate; and never hand money or credentials to an AI tool you cannot vet, given the surge in AI-enabled fraud. A general chatbot also cannot see your actual holdings unless you connect them, which is why bring-your-own-AI tools that read your real portfolio give more useful answers than a chatbot working from the internet's average.