EU Regulator: AI Poses Risks to Retail Investment Services

Europe’s markets regulator is warning about potential dangers of artificial intelligence (AI) use by investment firms.

The European Securities and Markets Authority (ESMA) issued guidance Thursday (May 30) for companies using AI when providing investment services to retail clients.

“When using AI, ESMA expects firms to comply with relevant MiFID II requirements, particularly when it comes to organizational aspects, conduct of business, and their regulatory obligation to act in the best interest of the client,” the regulator said, referring to Europe’s securities law.

ESMA said that while AI has potential benefits to firms and clients, the technology also brings with it risks, including algorithmic biases and data quality issues and “opaque decision-making by a firm’s staff members.”

The regulator also cautioned against over reliance on AI by both companies and clients for decision-making; and privacy “and security concerns linked to the collection, storage, and processing of the large amount of data needed by AI systems.”

“Potential uses of AI by investment firms which would be covered by requirements under MiFID II include customer support, fraud detection, risk management, compliance, and support to firms in the provision of investment advice and portfolio management,” ESMA said.

ESMA said it and the National Competent Authorities (NCAs) will continue to monitor the use of AI in investment services and the relevant EU legal framework to decide whether further action is warranted in this area.

These efforts are happening at a time when 100% of the CFOs PYMNTS has surveyed say their companies are using generative AI (GenAI), although how those firms are using it varies.

“All that were surveyed said they used the technology for at least three internal functions,” PYMNTS wrote earlier this week. “However, in most cases, we found those use cases were what our researchers categorized as routine and low-risk, such as summarizing information or accessing a catalog.”

Just 30% of surveyed enterprises are taking full advantage of their GenAI applications by deploying them in strategic ways, such as monitoring multiple processes, making assessments based on diverse inputs or creating new content.

“What is noteworthy, though, is that the firms using GenAI applications in highly impactful ways had not necessarily invested more initially,” that report said. “In fact, those firms with a high total impact score — meaning they used more strategic and complex GenAI applications — actually invested 25% less in the technology than those using GenAI for non-strategic applications.”

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.