June 8th 2025
This week’s highlights include the growth of AI-powered trading platforms, regulatory progress with BaFin adopting AI surveillance and new ISO standards, ongoing legal disputes over data rights, the transition from LLMs to LRMs in trading technology, global challenges around energy and connectivity for AI, and key developments from London Tech Week and the AI Summit.
1. AI Innovation in Trading Technology: Ingenuity Trading & FIX Nordics
After exploring the latest AI developments at FIX Nordics (read more here), I found another example of AI’s growing presence: Ingenuity Trading. With Naz Al-Khudairi (ex Credit Suisse AES, ex Barclays BARX) joining as Operating Partner alongside co-founders Pallav Chhaochhria and Karthik Teki, Ingenuity is bringing advanced algorithmic execution and AI to institutional investors through a multi-asset platform. Their integrated approach aims to enhance productivity, minimize workflow friction, reduce errors, and lower total execution costs—showing just how essential AI is becoming in institutional trading (read more). This trend parallels the recent announcement by Robinhood, which launched its advanced browser-based desktop trading platform, Robinhood Legend for retail investing.
2. AQR “Surrenders to the Machines”; AI in Research & Data Rights
After “years of scepticism,” AQR Capital Management’s Cliff Asness has embraced AI and machine learning more deeply across multi-asset strategies, boosting fund performance (Apex +19%, Delphi +14.6%). However, explainability in down markets remains a challenge.
AI-powered research is also accelerating: Perplexity’s integration with EDGAR now provides seamless access to all SEC filings for financial analysis, with new sell-side research and shareholder distribution features coming soon (details). Meanwhile, legal battles over AI training data continue: Reddit’s lawsuit against Anthropic for scraping user posts without permission highlights the rising tension over data rights, with OpenAI and Google opting to pay for access. The outcomes will shape future data ethics and ownership in AI research.
3. From LLMs to LRMs: Evolving AI Agents in Trading
The transition from LLMs (Large Language Models) to LRMs (Large Reasoning Models) is under way in trading technology. Today’s “agentic AI” acts as a careful assistant, not an autonomous trader—a key advantage where oversight and adaptability are essential. New agents built with frameworks like LangGraph and Codestral can generate, execute, validate, and refine strategies, automating workflows while still requiring human review. Apple’s new study notes LLMs excel at simple tasks but falter with complex, multi-step reasoning. LRMs, using advanced techniques like chain-of-thought prompting and RL, are more promising for sophisticated domains like trading—though even LRMs need human partnership (more).
4. AI Regulation & Standards: BaFin and ISO Developments
BaFin, the German regulator, has now integrated AI into its surveillance systems to detect suspicious trading patterns—a milestone in its post‑Wirecard reforms (Reuters). This follows major discussions on the future of the EU AI Act and a new international standard: ISO/IEC 42005 FDIS—AI Impact Assessment (ISO press release). This standard is a milestone for governance, offering a structured approach to assess/document AI risks and benefits. Join the next FIX AI Working Group meeting on June 17 for more on the Agentic Protocol; DM for details.
5. Energy, Compute & Geopolitics: AI’s Physical & Strategic Impact
AI’s explosive growth is reshaping global energy, compute, and connectivity needs (previous commentary). Oracle/OpenAI’s “Stargate” project is driving unprecedented demand for power and cooling (Financial Times), while Meta has signed a 20-year deal for an entire nuclear plant’s output (NY Post). Fiber optic cable shortages are emerging as a new bottleneck (Global Data Center Hub), and China’s export curbs on key materials add to global supply chain uncertainty (Financial Times).
6. London Tech Week & AI Summit: Eyes on the Future
It’s London Tech Week and AI Summit—with all eyes on Jensen Huang’s keynote and potential new NVIDIA investments. There’s speculation about fresh UK government AI initiatives (Gov.UK app, “sovereign AI”). Watch for speakers like Dr Ekaterina Abramova (LBS), Arunita Roy (FCA), Andy McMahon (Barclays), Jürgen Blumberg (Goldman Sachs), and Jamaria Kong (TowerBrook Capital Partners) (read more).
I will report back on what I learnt next week but as always, thanks for reading! Let me know what you liked, didn’t and what you would like to see more of.
Many thanks
Rebecca
