Skip to main content
Marsh logo

Article

AI risk management in a regulated landscape

Establish proactive risk management, adhere to compliance, incorporate ethics, foster collaboration, and adapt to regulations. Read this blog to learn more about AI risk management.
Abstract digital human face. Artificial intelligence concept of big data or cyber security. 3D illustration
Person's name and title

By Cihan Yildirim , 

Managing Consultant – Controls Assurance, Marsh Advisory

27/12/2023 · 3 minute read

The introduction of artificial intelligence (AI) has transformed the way we live and work. AI has the capability to solve complex problems, automate tasks, and enhance decision-making. However, integrating AI also creates a range of new risks and challenges that demand careful consideration. Organisations using AI must prioritise a robust risk management framework that aligns with evolving regulatory requirements.

Embracing high-risk AI compliance requirements

The AI regulatory landscape is rapidly evolving, with the European Union's AI Act and the United Kingdom's AI Regulation poised to become law in the coming years. The EU AI Act defines high-risk AI systems as AI systems capable of causing significant harm or posing risks to the fundamental rights and safety of individuals. For example, this would include AI systems used in critical sectors, such as transportation, energy, healthcare, and law enforcement where the potential risks are high. These regulations place significant emphasis on high-risk AI systems − posing stricter requirements for transparency, accountability, and risk mitigation. Therefore, organisations will be expected to conduct conformity assessments, maintain comprehensive records, and provide users with clear explanations of AI decisions. By complying with these stringent requirements, organisations can minimise the risk of non-compliance penalties and ensure that their AI systems operate in a responsible manner.

Prioritising proactive risk management 

In the absence of a comprehensive regulatory framework, it would be prudent to develop proactive risk management for responsible AI development and deployment in the future. Organisations must embed risk mitigation into every process involving AI. This enables organisations to identify potential risks related to bias, security, and privacy, while developing strategies to effectively mitigate these risks.

Ethical considerations

Ethical considerations are fundamental to responsible AI. Organisations must prioritise fairness, transparency, and accountability to ensure that AI systems align with societal values and respect fundamental rights. Considering these factors allows organisations to develop AI that prevents discriminatory outcomes and upholds ethical principles. The EU AI Act recommends the establishment of independent ethical review boards to provide expert advice and guidance on AI projects, ensuring that ethical considerations are embedded throughout the development process. Similarly, The UK AI Regulation requires organisations to conduct ethical risk assessments and establish ethical guidelines for their AI activities.

Collaboration and continuous adaptation

The AI regulatory landscape is constantly shifting, requiring organisations to be agile and adaptable. Collaboration between regulators, industry leaders, and academic experts, is crucial to staying informed about industry developments, shaping the future of AI regulation, and sharing best practices. Organisations must also continuously update their risk management practices to align with evolving regulatory standards − ensuring that their AI systems remain compliant. This collaborative approach fosters innovation while ensuring that AI development and deployment adhere to ethical principles and societal expectations.

Striking a balance between innovation and responsibility

Successful AI risk management ensures that AI is developed and deployed responsibly. It is crucial that organisations allow themselves to take advantage of the opportunities AI can offer. To do this while navigating the complex AI risk landscape, it is important organisations do the following:

  • Establish proactive risk management. 
  • Adhere to stringent compliance requirements.
  • Incorporate ethical considerations, where possible.
  • Foster a collaborative environment. 
  • Continuously adapt to evolving regulatory standards. 

To find out more about how to establish a risk management framework surrounding AI for your business, reach out to an adviser. 

Related insights

Business meeting in the wide modern office with stock infographics

Article

How risk committees can help navigate complex business environments

04/03/2024
Valuable business documents including contracts. Generative AI.

Article

Could AI be the solution to companies’ supply chain conundrum?

27/02/2024
Stock market hologram, blue and violet forex hud with numbers and lines dynamics. Concept of diagrams, online trading and finance. 3D rendering illustration

Article

The economic climate's impact on insurance providers’ balance sheets

06/02/2024
Down angle view of the city skyscrapers.

Video

Property Reinstatement Valuations

19/01/2024
View more