EU: Consultation on the use of Artificial Intelligence (AI) within financial services

Why should I read this?

The European Commission has launched a consultation to gather insights on the use of Artificial Intelligence (AI) within the financial sector, “Targeted consultation on artificial intelligence in the financial sector”.  This initiative is part of the Commission’s broader strategy to ensure that AI development aligns with EU values and regulations, particularly the AI Act, which comes into force in July 2024.

AI was in use within financial services long before public discourse on AI and works on the AI Act began.  The consultation aims to identify key use cases, benefits, barriers and risks associated with AI applications in financial services.  Stakeholders are encouraged to share their experiences and perspectives to inform the Commission’s approach to AI governance in financial services. 

Mindful that financial services firms are concerned about duplicate regulation, the Commission is consulting for the expressly purpose of avoiding this.  The Commission’s stated goal is to harmonise policy with existing financial services regulations like MIFID/R and market abuse rules, to avoid stifling innovation.

Example use cases and risks

The consultation identifies the following possible use cases on which it seeks insights:

• Fraud detection
  • Data analysis to detect patterns and anomalies that may denote fraud
• Risk management
  • Analysing and predicting market trends, credit risks and investment opportunities
• Automation of routine tasks
  • Data entry, transaction processing, and document verification
• Cost savings
  • Automating processes and improving efficiency
• Personalised financial advice
  • Robo advisors
• Compliance and regulatory support
  • Interpreting regulatory requirements and monitoring transactions
• Enhanced decision-making
  • Deriving insights from data analysis to inform decisions
• Improved security
  • Identifying potential security threats, detecting unusual patterns of behaviour and providing real-time alerts
• Streamlined processes
• Improved customer service
  • Personalised and efficient customer service, eg chatbots

The consultation identifies the following possible risks on which it seeks insights:

• Lack of access to the required data, in general
• Lack of access to the data in an appropriate digital format
• Lack of access to appropriate data processing technology
• Data privacy
• Lack of trust in the technology
• Regulatory compliance with financial regulation:
  • Not all types of AI applications are in line with the requirements of financial services regulations
• Innovation:
  • Combining AI with other technologies to enhance its potential and generate new services
• Transparency and explainability:
  • Difficult to explain, complex and opaque AI algorithms can create issues of trust and accountability.
• Bias and discrimination:
  • Arising from bias in the data used to train AI models
• Reputational risk from undesirable AI behaviour or output
• Liability risks:
  • Legal uncertainty as to who bears the liability for losses and damages caused by malfunctions of AI applications
• Shortage of skills to develop AI
• Dependability of AI
  • The problem worsens as financial services firms become more reliant on AI
• Job displacement:
  • Financial services roles automated by AI may become redundant
• Cybersecurity:
  • AI vulnerability to cyber attack
• Integration challenges:
  • Integrating AI technologies with existing systems can be complex and expensive
• Additional costs of AI:
  • Acquiring or developing applications, keeping them up to date, training staff

Next steps

The consultation is open for responses in the form of an online questionnaire until 13 September 2024. The Commission plans to publish a report summarising the findings, which will play a role on shaping future EU regulation of AI in financial services.

How Eversheds Sutherland can help

Our team is experienced in AI ethics and can work with you to develop an AI governance framework, together with accompanying training material, tailored to your sector and use of AI. We also offer training for firms looking to bring AI capability in-house, utilising members of our Konexo team, to deepen understanding of the issues from both a technical and legal perspective.

We work closely with financial services clients around the world to understand the thematic shifts and surrounding regulatory environment shaping products, processes and business models. We boast a multi-disciplinary, multi-jurisdictional team which is at the forefront of change across the global themes of AI, Biometrics, Blockchain and crypto, cloud, cybersecurity, digitisation, payments, operational resilience and Regtech and which transcends traditional areas of practice.