The Promise and the Question
If you have opened a financial app, read a fintech newsletter, or simply scrolled past an investment advertisement in the last five years, you have encountered the pitch: artificial intelligence is transforming investing. Algorithms make faster decisions, have no emotions, never panic-sell, and charge a fraction of what a human adviser costs. Why, the argument goes, would anyone trust a person when a machine can do the same job more efficiently?
It is a compelling story. And like most compelling stories in finance, it contains important truths alongside important omissions.
AI and algorithmic tools have genuinely changed what is possible in portfolio management. They have democratised access to investment products that were once only available to wealthy clients. They have driven down fees significantly. And in certain specific tasks, they perform better than any human. But the picture is more complicated when you look at where AI genuinely adds value, where human judgment remains essential, and what the actual performance data shows across different market conditions.
This article is an honest look at both.
What We Actually Mean by AI in Investing
The term AI in investing covers a broad and often misunderstood range of tools. Before comparing performance, it helps to understand what each category actually does.
Robo-advisors
Robo-advisors are the most widely known and widely used form of AI-driven investing for individual investors. Platforms like Betterment, Wealthfront, Vanguard Digital Advisor, Nutmeg in the UK, Scalable Capital in Europe, Wealthsimple in Canada, and StashAway across Southeast Asia use algorithms to build diversified portfolios of ETFs based on an investor's stated goals, time horizon, and risk tolerance.
According to Statista, global robo-advisor assets under management reached approximately $1.97 trillion in 2025, with the United States accounting for around $1.57 trillion of that total. The market is growing rapidly, with Morningstar's analysis putting robo-advisor AUM between $634 billion and $754 billion in the US retail market alone in 2024. Asia-Pacific is currently the fastest-growing region, driven by regulatory sandboxes in Singapore, Japan, and India that have accelerated adoption.
The core value proposition of robo-advisors is straightforward: low fees, diversified portfolios, automatic rebalancing, and accessible minimum investment thresholds. Morningstar's 2024 report on digital advice found the median robo-advisor fee was 0.25% of assets per year, compared to around 1% for a human financial adviser. That fee gap, compounded over decades, is genuinely significant.
Algorithmic and quantitative trading
At the institutional level, AI-driven investing is considerably more sophisticated. Quantitative hedge funds like Renaissance Technologies, Two Sigma, and Man Group use machine learning models that analyse vast datasets, including alternative data sources like satellite imagery, credit card spending patterns, and earnings call transcripts, to generate trading signals. These systems execute trades in milliseconds and can identify patterns that no human analyst would detect.
Bloomberg Intelligence data indicates that AI-driven quantitative strategies contributed more than 40% of hedge fund trading volumes in 2024. These are not simple rule-based algorithms; they are adaptive systems that continuously update their models as market conditions change.
Generative AI as an advisory tool
The newest frontier involves large language models being deployed to assist human advisers rather than replace them. Morgan Stanley has integrated OpenAI's technology, with 98.5% of its wealth management teams using the AI assistant at least once per week as of 2024. JP Morgan is testing a generative AI system within its private banking arm. The goal is not to automate away advisers but to make them faster and more informed, capable of answering client questions in real time rather than following up after research.
The Performance Evidence: Where AI Wins and Where It Does Not
This is where the conversation needs to be honest and specific, because the answer depends heavily on what type of investing you are talking about and under what market conditions.
Robo-advisors: competitive in calm markets, comparable but not superior overall
For standard portfolio management in a diversified multi-asset framework, robo-advisors have demonstrated that automated management can generate competitive returns. According to Bankrate's analysis of Condor Capital's Robo Report, robo-advised portfolios with a 60/40 stock-bond allocation produced average annual returns of 7% to 9% over the five years ending September 2024. Those returns are broadly comparable to what human-managed diversified portfolios have delivered over the same period.
Betterment has reported that its portfolios have outperformed the average investor 88% of the time over the past decade. This is not primarily because the algorithm is clever. It is because robo-advisors impose the discipline that most individual investors lack: they do not panic-sell during drawdowns, they rebalance automatically when allocations drift, and they apply tax-loss harvesting consistently rather than selectively.
That last point matters enormously. The Financial Conduct Authority in the UK found that fully automated robo-advisors reduced investment management fees by an average of 65% compared to traditional advisory services in 2023. On a large portfolio held over many years, that fee reduction alone can be more valuable than incremental performance differences.
AI funds in volatile markets: a genuine edge on the downside
Academic research published in Springer's Future Business Journal examined AI-driven funds versus human-managed funds across distinct market cycles from 2022 to 2024. The findings were notable: in the 2022 bear market, AI-driven funds had a mean return of negative 17%, compared to negative 30.7% for human-managed funds. AI systems' systematic, rules-based approach to risk management, including algorithmic hedging and stop-loss mechanisms, provided meaningfully better downside protection.
However, the pattern reversed in recovery periods. In the 2023 to 2024 bull market, human-managed funds outperformed their AI counterparts, because human managers were better able to apply qualitative judgment in identifying early-stage opportunities that models trained on historical data had not accounted for.
This pattern points to an important structural truth: AI tends to do better in conditions it has been trained to recognise, including volatility regimes and known risk patterns. Humans retain an edge in genuinely novel situations that fall outside the historical data distribution on which AI models were built.
Quant hedge funds: early promise, declining edge
At the elite institutional level, the story of AI-driven investing is increasingly one of initial advantage eroding over time. Research published by Alpha Architect, drawing on academic analysis of AI fund performance across multiple market cycles, finds that early AI funds did generate significant excess returns (alpha) in their early years. However, that outperformance declined consistently as more funds adopted similar strategies, traded on correlated signals, and competed for the same mispricing opportunities.
Renaissance Technologies' Medallion Fund remains the most famous exception: its extraordinary long-term returns have demonstrated what proprietary AI-driven trading can achieve in the hands of the right people with the right data. But Medallion is closed to outside investors and operates at a scale and with a level of secrecy that makes it essentially irrelevant as a benchmark for ordinary investment decisions.
The broader quant fund world tells a more sobering story. When AI strategies proliferate, the market adjusts, the signals decay, and the alpha disappears. As Larry Swedroe of Alpha Architect has noted, the competition is not between an AI and an individual human: it is between an AI and the collective decision-making of millions of market participants, many of whom are also running AI models. That is a significantly harder problem.
What AI Cannot Do
Understanding where AI adds genuine value requires equal clarity about where it does not.
Complex, whole-life financial planning
Robo-advisors are investment management tools. They are generally not financial planning tools in the full sense. A platform that builds you a diversified ETF portfolio does not know that you are planning to sell your business in three years, that your spouse is in a defined-benefit pension scheme, that you have a child with a disability whose care needs to be funded into perpetuity, or that your estate plan has not been reviewed since a change in tax law. These considerations require a human who understands your whole situation and can integrate investment decisions with tax planning, succession planning, insurance, and estate structuring.
Bankrate describes a "ceiling of complexity" beyond which robo-advisors cannot go. Below that ceiling, they are highly effective. Above it, the limits of purely algorithmic advice become apparent quickly.
Behavioural coaching during genuine crises
The 2020 COVID crash and the 2022 rate-shock bear market both demonstrated something that no model had fully anticipated. In those moments, the value of a human adviser is not analytical; it is emotional. The ability to sit with a client who is frightened, explain what is happening in context, discourage panic-selling, and maintain the long-term plan requires a quality that current AI systems cannot replicate: genuine empathy and relational trust built over time.
Adapting to genuinely novel circumstances
AI models learn from historical data. When market conditions move beyond the range of what the training data contains, as they did during the early pandemic period when central banks deployed unprecedented policy interventions, the models can fail in unexpected ways. Renaissance's institutional funds, which are designed differently from the Medallion Fund, declined significantly in 2020 because their models were calibrated to normal market behaviour, not to the specific dynamics of a pandemic-driven crash followed by an extraordinary policy rebound. Human managers who understood the mechanism, not just the historical patterns, navigated that period better.
The Hybrid Model: Where the Industry Is Actually Heading
The most honest answer to the question of AI versus human advisers is that the framing itself is increasingly obsolete. The industry is moving toward models that combine both.
Institutional Investor research finds that personalisation remains the critical differentiator that AI has not yet displaced, and that large institutions including Goldman Sachs, Morgan Stanley, and JP Morgan are investing heavily in AI not to replace human advisers but to make them faster and more capable. The model that is emerging is one where AI handles what it does well, data analysis, portfolio construction, rebalancing, tax optimisation, and pattern recognition, while humans focus on what they do well: understanding clients, navigating complexity, providing emotional steadiness, and integrating investment decisions with the full range of a client's financial life.
Vanguard's research on the value of human advice has estimated that working with a skilled financial adviser can add approximately 3% per year in net returns through a combination of behavioural coaching, asset allocation, tax efficiency, and financial planning. Not all of that is attributable to superior stock selection. Much of it comes from preventing the behavioural mistakes that investors make without guidance, from panic-selling during downturns to underinvesting during accumulation.
The Morningstar analysis of the best robo-advisors notes that the most successful platforms are those that have added human touchpoints alongside their automated core, whether through access to certified financial planners at scale, hybrid models that blend algorithmic portfolio management with human relationship management, or tiered service structures that route simpler accounts to automation while keeping complex clients with dedicated advisers.
A Global Picture: How AI Investing Is Playing Out Across Markets
The adoption of AI-driven investing tools is not uniform globally, and the regulatory and market context shapes how effective these tools are in different regions.
In the United States, robo-advisors are well-established and regulated under existing SEC frameworks. The market is competitive, fees have been driven down considerably, and the major platforms have track records long enough to evaluate. North America accounted for around 38% of global robo-advisory revenue in 2024, according to Mordor Intelligence.
In Europe, MiFID II regulation has imposed transparency requirements that have actually benefited investors using digital advisory platforms, by mandating clear disclosure of costs and conflicts of interest. Pan-European platforms like Scalable Capital have grown rapidly, and the European Securities and Markets Authority has issued guidance on AI use in financial services that is creating a more standardised compliance environment.
In Asia-Pacific, the growth story is the most dramatic. Singapore's Monetary Authority, Japan's FSA, and India's SEBI have all created regulatory frameworks that allow digital investment platforms to operate and experiment within defined guardrails. SEBI data from 2024 found that 28% of first-time retail investors in India now prefer robo-advisory platforms. Platforms like StashAway and Syfe are expanding rapidly across Southeast Asia. China's Ant Group has added ESG-focused algorithmic portfolios to its wealth management offerings. Asia-Pacific is projected to grow at a compound annual rate of more than 30% through 2030.
In emerging markets across sub-Saharan Africa and Latin America, the opportunity for AI-driven investing tools is significant precisely because human financial advice has historically been accessible only to the wealthy. Lower-cost digital platforms have the potential to extend real investment management to populations that have never had access to it before, a genuinely meaningful development that goes beyond the performance debate.
What This Means for Your Portfolio
For most individual investors, the practical implications of this debate are clearer than the theoretical arguments suggest.
If you are at the beginning of your investment journey, have a relatively straightforward financial situation, and primarily need disciplined, diversified, low-cost investment management, a well-designed robo-advisor platform is a genuinely good option. The fee savings are real, the discipline is built in, and the access to diversified market exposure is meaningful.
As your wealth grows, your financial situation becomes more complex, and the stakes of individual decisions rise, the value of skilled human advice grows proportionately. The questions that matter most at higher levels of wealth, including tax optimisation across jurisdictions, estate planning, business succession, pension structuring, and generational wealth transfer, are not questions that any current AI platform can answer adequately.
The most sophisticated investors in the world are not choosing between AI and human advisers. They are using both, allocating each to the tasks it is actually suited to, and building relationships with advisers who are themselves augmented by the best available technology.
How Celerey Thinks About This
At Celerey, we do not think the question is whether to use technology or human judgment. We think the question is how to combine both intelligently so that our clients benefit from cost efficiency, analytical rigour, and the kind of personalised, whole-life advice that no algorithm currently provides.
We use the best available data, analytical tools, and portfolio construction frameworks to inform the advice we give. And we bring human expertise, relationship, and judgment to the decisions that genuinely require it. Tax planning, estate structuring, business transition planning, and navigating major life events are not tasks you want to delegate to an algorithm, however well-designed.
If you are wondering whether your current investment setup is genuinely working for you, whether it is robo-managed, human-managed, or some combination, that is exactly the kind of conversation we are here to have. Reach out to the Celerey team to begin a portfolio review.