AI Summary
5 min readThe 86x Token Explosion: Inside Man Group's AI Transformation
In just six months, Man Group's token consumption grew 86 times. That is not a typo. Since January, the world's largest publicly traded hedge fund has seen its use of AI models explode across every part of the firm—not just the quant teams, but finance, operations, and even the people team. "We were not expecting usage to be that high," says Tushara Fernando, the firm's head of data and AI. The growth is so steep that the firm hasn't yet bothered to build a router that would automatically send queries to cheaper models. They are still in the experimentation phase, letting people figure out what works.
The Two Kinds of AI in Asset Management
Fernando draws a clear line between traditional machine learning and generative AI. Machine learning in quant finance has long meant linear regressions and neural nets that predict a feature—does this stock look cheap? Does this signal predict returns? Generative AI is different. It is about enablement: creating things, synthesizing information, and augmenting human judgment rather than just predicting a number.
Continue reading the full summary in the app — free to try.
Read Full Summary →Free • No credit card required
Never miss an episode of Odd Lots
Get every new episode summarized in your inbox — free, ~5 minutes to read.
No spam. Unsubscribe anytime.
What you'll learn
- 1 (01:35) **Introduction: The AI-in-Investing Puzzle** - Joe and Tracy frame the core question: is generative AI a minor evolution of quant/ML or something more revolutionary?
- 2 (05:52) **Guest Introductions: Gary Collier (CTO) and Tushara Fernando (Head of Data & AI)** - They describe Man Group as a "full spectrum active asset manager" and their roles spanning technology, data, and AI.
- 3 (07:32) **Machine Learning vs. Generative AI** - Tushara explains the shift from traditional ML (regressions, neural nets for prediction) to generative AI (enabling creation and augmentation).
- 4 (08:56) **What Has Changed on the Trading Floor Since 2022** - Gary describes how AI tools are now visible on every desk, augmenting all roles.
- 5 (13:08) **Agentic Workflows: From Research to Execution** - Tushara explains that agentic systems are not a future concept; they are already in production.
- 6 (18:18) **The Real Bottleneck: Organizational Change, Not Compute** - Gary identifies the bottleneck as the human capacity to safely deploy AI at scale in a regulated environment.
- 7 (20:53) **The Data Stack: Three Layers** - Tushara breaks down the data architecture: structured market data (tick data, terabytes), unstructured alternative data (messy, needs tagging and connection), and institutional knowledge (a new layer for teaching AI about Man Group's processes).
+ Full timestamped outline available in the app
Show Notes
We've gone through a number of a technological revolutions in investing, whether it was the dawn of the high frequency trading era or the introduction of robotraders. When it comes to AI, the big question that remains in the investment context is whether or not the technology will be implemented like those past tech innovations — meaning it will be integrated into the flow of the business without upending everything as we know it — or if AI will transform the very nature of investing. Right now, AI's use in investing is a mixed bag: People are excited about its potential, but several firms are still trying to figure out its value. Today, we speak with Man Group's CTO Gary Collier and Head of Data and AI Tushara Fernando about how one of the largest publicly-traded hedge funds in the world is actually implementing AI into its work. We speak with them about empowering their quants with AI tools, the challenge of integrating AI safely, and the creative ways their staff is thinking about token spending, which is up 86-fold this year.
Only Bloomberg.com subscribers can get the Odd Lots newsletter in their inbox — plus unlimited access to the site and app. Sign up here: bloomberg.com/subscriptions/oddlots
See omnystudio.com/listener for privacy information.
More from this podcast
Odd Lots →