Not by a discretionary fund manager with a Mayfair address, but by an app promising machine-learning portfolios and automated trading. The pitch lands well with people who automate everything else in their businesses. Whether the technology deserves the trust is a different question.

Why business owners are the target market

The AI investing pitch is aimed squarely at the time-poor and cash-rich, which describes most owner-directors. The reasoning feels sound: you automated your invoicing, your marketing and half your customer service, so why not the money? Subscription pricing feels familiar. Dashboards feel like management information. And the promise of removing emotion from investment decisions appeals to anyone who has watched their own judgement wobble during a bad quarter.

The scale of the shift is measurable. UK trading statistics drawn from FCA data show retail participation in markets at record levels, with automated and app-based trading among the fastest-growing segments – growth that includes a striking number of accounts funded from business income.

What the algorithm actually signed up to

Most products sold as AI trading fall short of the label. Some are screeners that suggest trades a human still approves. Many are rule-based systems executing strategies that predate the microchip, rebadged for the moment. A minority genuinely adapt to market conditions. None of them, crucially, carries a fiduciary duty. The software has no obligation to act in your interest, no qualification to check, and usually no regulation of its own – the regulatory protection sits with the broker where the trades execute, not the algorithm deciding them.

That distinction matters for directors more than most, because the sums are larger and the tax wrappers less forgiving of error. An algorithm churning a general investment account can generate a capital gains headache no dashboard will warn you about.

Due diligence, the business version

The sensible approach is the one owners already apply to suppliers: independent verification before signing. Reviews of the best AI trading app options for UK users now test these systems with real funded accounts, measuring forward performance, true costs and whether the automation claims survive contact with a live market. The results are sobering: vendor backtests and real-world outcomes routinely part company, and the gap is usually the subscription fee plus a good deal more.

Independent researchers such as The Investors Centre, which deposits its own money to test platforms, consistently find that venue selection – the regulated broker underneath the software – moves outcomes more than the choice of algorithm on top.

The delegation test

There is a version of this that works. Automated investing with modest claims, regulated execution and costs you have actually totted up can suit a busy director well. The test is the same one you would apply to any hire: demand evidence of past performance in live conditions, understand exactly what it costs, know who is accountable when it goes wrong, and start with a probation period. If an applicant refused all four, you would not give them the company chequebook. The algorithm should not get the pension on easier terms.

Read the full article →