- Free report: Multi-managers: Key trends in technology, operations, platform growth
- Join the AFI Private Markets Wealth and Innovation Summit in Mayfair, 25 June 2026
Artificial intelligence is rapidly reshaping quantitative investing, with asset managers increasingly using machine learning, natural language processing and generative AI to enhance stock selection, portfolio construction and research efficiency.
Speaking on BNP Paribas Asset Management’s Talking Heads podcast, Ram Rasaratnam, CIO for Quant Equity Strategies at AXA IM Core, part of BNP Paribas Asset Management, outlined how AI is becoming embedded across the firm’s investment process as managers search for incremental performance gains in increasingly competitive markets.
Rasaratnam, whose team recently received industry recognition for quantitative investing and AI adoption, said the group has been using machine learning techniques for almost a decade. One early neural-network model designed to predict short-term volatility spikes successfully identified risks in Silicon Valley Bank and Signature Bank ahead of the US regional banking crisis in 2023.
AI moves deeper into stock selection
The firm is also using natural language processing to analyse corporate earnings-call transcripts, helping quantify management sentiment, confidence and communication changes over time.
Rather than replacing traditional factor investing approaches, Rasaratnam said AI is enhancing existing models focused on valuation, earnings quality and growth.
“What we’ve seen to stay competitive and keep an edge is really about marginal gains,” he said. “We’re not changing fundamentally our approach. We’re just trying to find these new data sets to create those marginal gains.”
The growing use of generative AI tools is also accelerating research workflows. Rasaratnam highlighted the firm’s use of internally hosted large language models capable of analysing vast repositories of proprietary code, answering natural-language queries and even generating code that conforms to the firm’s internal standards.
He described the technology as a “game-changer” for productivity, allowing researchers to spend less time writing code and more time interpreting outputs and designing models.
Human oversight still critical
Despite the rapid adoption of AI, Rasaratnam stressed that human judgement remains essential, particularly given the complexity and constantly evolving nature of equity markets.
“Equity markets are incredibly complex, they’re not stationary,” he said. “We still need that human judgment and the expertise that our team have.”
Governance has become increasingly important as firms integrate AI into investment systems. Rasaratnam said all AI-generated code undergoes extensive testing and requires sign-off before entering production environments.
The comments come as hedge fund and asset management executives increasingly prioritise AI-related skills in recruitment. Investment leaders at firms including BlackRock, Man Group, Point72 Asset Management and AQR have highlighted growing demand for quantitative analysts with expertise in machine learning, data science and AI-assisted research. They seek quants who can effectively apply AI techniques to financial modelling and risk assessment. Additionally, proficiency in programming languages like Python and R remains crucial, as these tools are integral to developing and implementing AI-driven strategies
Beyond technical capabilities, firms are increasingly looking for investors who can effectively collaborate with AI systems while maintaining traditional financial and risk-management discipline.
Rasaratnam also suggested AI is allowing quant managers to move into areas historically dominated by fundamental investors. One emerging area of research involves analysing patent filings using machine learning to assess corporate innovation and future profitability.
“Ten years ago, we wouldn’t have been able to build insight from that type of data,” he said. “Now we can challenge more traditional approaches.”
“Ten years ago, we wouldn’t have been able to get into that space and build insight with that data, whereas now we can,” he said. ” And I think there’s other data sources out there that allow us to challenge more traditional approaches.”