A new survey from Substantive Research covering 50 asset managers with $20 trillion in assets shows global research budgets were nearly flat in the first half of 2026, rising only 0.4% in monetary terms, even as spending on AI-enabled research analytics and tools jumped 40% over the previous year. AI tools now take about 7% of the average research budget, redistributing spending away from investment banks, specialist research houses and independent analysts rather than creating new revenue for traditional providers.
According to the data, the 10 largest research providers receive 56% of the average institutional research budget, while the three largest account for 23%. Independent research providers receive only 8%. North American research spending fell 6% as a proportion of assets under management, while European spending rose just over 1%; both remain more than 50% and 60% below 2018 levels respectively, reflecting the long shadow of MiFID II research-payment reforms. The UK Financial Conduct Authority's PS24/9 policy has introduced more flexibility, but adoption has not yet produced a material spending increase. Substantive Research CEO Mike Carrodus said the central commercial question is whether delivering research into buy-side large language models represents another distribution channel or a distinct use of intellectual property that warrants additional payment.
Separately, Money20/20's Chief Strategy and Growth Officer Scarlett Sieber said in an interview that AI agents in finance remain years from meaningful scale. She noted a shift from exploring use cases to safe deployment, saying roughly 30% of speaker submissions for Money20/20 US included the words AI and fraud. Sieber traced the current focus back to earlier experiments such as Bank of America's Erica chatbot and argued that interoperability remains a major weakness, but stablecoins could help connect financial institutions, especially for cross-border payments. She cited a consortium of 21 financial institutions exploring a stablecoin company and said the intention to use a public blockchain was significant.
Sieber sees stablecoins becoming the most likely key piece for interconnectedness and possibly the default for AI-agent transactions, with growing institutional interest alongside AI, payments, digital assets and identity infrastructure. However, she said truly scalable autonomous finance is still a few years away because of fragmented regulation, accountability challenges and the need to prove consumer benefits. The combined picture is one of technological redistribution and early-stage integration: AI is changing how financial information is consumed, while stablecoins are increasingly discussed as the payment layer for a future wave of autonomous agents, though no specific cryptocurrency was identified as a direct beneficiary in either report.
