AI levelling the investment field
Artificial intelligence is fast becoming one of the most powerful forces reshaping global finance.
At the annual Asian Financial Forum held earlier this year in Hong Kong, leading industry voices painted a picture of a very near financial future driven by artificial intelligence (AI), where algorithms are rapidly surpassing humans not just in speed, but in the capacity to analyse, synthesise and act on data.
“Generative AI is the single most disruptive technology that we have ever experienced in human history,” said Sinovation Ventures chair and AI expert Dr Kai-Fu Lee.
“We now have AI thinking better and faster than people most of the time for most tasks.”
Lee argued that this shift is not limited to trading desks or research teams, but that every department in a financial firm should be incorporating AI tools.
He pointed out that the number-centric nature of finance makes it especially conducive to fast, scalable deployment of AI.
“You can’t use AI to make a car instantly, but in the financial industry you are not shipping physical goods, you’re dealing with numbers.”
Lee, formerly the head of Google China, isn’t talking about a hypothetical future.
He cited an AI-enhanced market index fund backed by his venture capital (VC) firm, which allows only AI to buy and sell stocks – humans are excluded from the process entirely. He said the fund outperforms the market index by around 30% each year.
High-Flyer Capital Management, a Chinese hedge fund founded in 2016, gained attention for using machine learning to identify mispriced stocks and time trades.
Its funds have returned 151% in total (or around 13% annualised) since 2017 – a standout performance amid a volatile China market. Regulatory changes in 2024 forced the closure of its market-neutral funds, but High-Flyer’s successes continue to influence a new wave of AI-led investment innovation.
Lee described the global AI race as a tale of two superpowers: the US, leading on groundbreaking research through its culture of innovation and strength in fundamental science, and China, excelling in the practical implementation of user-facing applications.
“WeChat is better than WhatsApp. TikTok is better than Instagram,” he noted. “Chinese teams have figured out how to find product–market fit globally.”
China’s fintech firms, in particular, have been early adopters of large language model (LLMs) like DeepSeek, which received attention earlier this year for claiming performance comparable to OpenAI’s GPT-4 at a significantly lower training cost.
Lee’s message to a room full of finance professionals was direct: if your firm is not integrating AI into research, trading and operations today, it is already falling behind.
“AI should be doing most of the writing. AI should be doing most of the reading. I use AI to read all my news … to ask what the top news are today, or what are three stocks I should buy or sell,” he said.
AI removes one of the most common pitfalls for investors: emotion. Tools now allow for real-time sentiment tracking and automated triggers based on logic and data.
But not every role will disappear. Lee sees long-term investing, M&A (mergers and acquisitions) and relationship-based advice remaining human-led. What’s at risk are roles driven by short-term analysis and repeatable decision-making.
“Computer trading replaced floor traders. AI trading will replace a lot of traders today,” he says.
“So now would be a good time for people in the financial industry to upgrade their skills. Otherwise, their job will simply be replaced.”
He suggested financial services firms consider appointing a chief AI officer: someone who understands the technology deeply and can lead transformation across departments, from legal and HR to asset management.
Democratising financial access
AI has already been embedded across the capital markets landscape, with large and small financial institutions using AI not just to assist human analysts, but to automate decision-making at scale. It is powering everything from trade execution and risk modelling to real-time sentiment tracking, portfolio optimisation and fraud detection.
But Lee says this is only just the beginning.
Until recently, building sophisticated AI models required hundreds of millions of dollars in computing power. Now, models are being trained for a fraction of that cost, enabling a much broader range of financial firms to implement AI technologies.
“AI will be made available to everyone – the world will be able to build applications on top,” said Lee.
For everyday investors, this level of access may prove to be one of the most transformative aspects of AI in finance. Previously, obtaining high-quality investment advice and in-depth data analysis meant relying on costly human advisers or institutional-grade tools that were beyond the reach of most individuals.
With AI-powered market analysis assistants emerging to bridge that gap, users will be able to query the markets in plain language, analyse stock trends in real time and receive suggestions tailored to their investment preferences.
Ultimately, it is expected that these assistants will offer tailored guidance to an individual’s specific profile, factoring in things like risk tolerance and income level, but also personal values and unique financial goals.
In some markets, AI regulators have already approved AI platforms for public use. But as these tools begin offering recommendations that resemble traditional financial advice, they raise important regulatory questions around licensing, disclosure, complaint processes and duty-of-care obligations.
Financial regulators around the world are looking at how to address these issues. With the right policy and regulatory frameworks in place, AI could help democratise investing, making smart, data-driven decisions accessible to all, not just the already wealthy.
– Tim McCready was a guest of the Asian Financial Forum