Compliance
On 30.05.2024, ESMA published a publication in which the supervisory authority addresses the topic of artificial intelligence. You can access the publication here.

Use of Artificial Intelligence - ESMA Guidelines

Summary of the contents of the supervisory authority's recommendation
This summary summarizes the main points of the ESMA document "Public Statement on the Use of Artificial Intelligence (AI) in the Provision of Retail Investment Services" (ESMA35-335435667-5924) and highlights the key aspects and recommendations.

Introduction:
The document1 from ESMA (European Securities and Markets Authority) deals with the use of artificial intelligence (AI) in the area of retail investment services. It highlights both the opportunities and risks associated with AI and provides recommendations on how investment firms can use AI in accordance with the requirements of MiFID II (Markets in Financial Instruments Directive).
Potential applications and benefits of AI
Customer service and support:
AI-powered chatbots and virtual assistants can answer customer queries and provide account information.

Investment advice and portfolio management:
AI tools can analyze large amounts of data to provide personalized recommendations and manage portfolios.

Compliance:
AI can help analyze financial regulations, compare internal policies and generate compliance reports.

Risk management:
Assess and monitor the risk of investments and portfolios.

Fraud detection:
Monitoring of transactions and communications to detect fraud.

Operational efficiency:
Automation of tasks such as data entry and report generation.

Risks for companies and customers
Excessive reliance on AI:
Lack of human judgment when making complex decisions.

Lack of transparency and explainability:
Difficulty in understanding AI decisions.

Security and data protection risks:
Concerns regarding the collection and processing of large amounts of data.

Reliability of results and data quality:
Erroneous or biased data can lead to incorrect results.

MiFID II requirements
Acting in the interests of the client:
Unchanged obligation, regardless of the technologies used.

Transparency:
Companies must disclose how AI is used in decision-making processes.

Organizational requirements:
Management should have an understanding of AI applications and ensure appropriate oversight.

Risk management:
Companies should have robust structures in place to identify and mitigate AI-related risks.

Employee training and competence:
Regular training on the risks and possible uses of AI.

Behavioral requirements
Compliance with MiFID II requirements:
Particularly important when using AI for investment advice and portfolio management.

Quality assurance:
In-depth testing and regular reviews of AI algorithms.

Data protection:
Compliance with data protection regulations to protect sensitive customer data.

Record-keeping obligations:
Investment firms must keep comprehensive records of the use of AI and related client complaints to ensure compliance with MiFID II.

Conclusion and next steps:
ESMA emphasizes the need to comply with MiFID II and prioritize client interests when using AI. Further resources and coordination with regulators is recommended. ESMA and national authorities will continue to monitor developments in AI use and the regulatory framework and take further action as appropriate. [/p]
1 ESMA35-335435667-5924

05/2024


At this point, we refer to the general rules of conduct for the use of applications that use artificial intelligence.