AI Dominates Institutional Research, But Energy Constraints Loom

Institutional investors have spoken, and AI is now their daily go-to for research. In a recent survey of 410 fixed-income investors, a powerful 52% of long-only managers and asset owners revealed they primarily use AI for research. Meanwhile, hedge funds, those old masters of market maneuvering, boast a staggering 72% daily AI usage. Humans still call the shots, but the bots are crunching the numbers with gusto.

But not all is smooth sailing in the AI waters. While AI might be the darling of research, it's a wallflower in trading and execution. Security concerns top the list of barriers, which keeps AI sidelined. Oh, and don't hold your breath for job cuts. Only 7% of respondents foresee any significant reduction in staff, so the revolution won't be televised in pink slip form.

Here's the thing. While AI's dominance in institutional research is undeniable, Marc Andreessen's take on energy and cooling constraints throws a wrench in the works. His AI:AC Hypothesis argues that AI's growth will hitch its wagon to available energy and air conditioning. With the IEA predicting data center demand to double by 2030, the infrastructure had better be ready. In the U.S., data centers might soon out-consume entire industries like aluminum and steel. If power and cooling don't keep up, AI's meteoric rise might hit a ceiling faster than anyone expects.

So, who wins and who loses? Institutions already deep in AI adoption are ahead of the pack, but the real winners will be those who can manage energy resources effectively. With major tech firms planning to pump a combined $725 billion into AI by 2026, the grid's ability to handle this demand will shape the future of AI, and maybe even your portfolio.