AI Risks and Global Compliance Strategies

AI Risk and Approaches to Global Regulatory Compliance

Introduction

As jurisdictions worldwide rapidly introduce Artificial Intelligence (AI) regulation, understanding the associated risks becomes a significant challenge for both regulators and organizations. This comprehensive analysis by Deloitte's Internet Regulation team aims to support companies in navigating these challenges.

Key Takeaways

The AI regulatory landscape is evolving rapidly, with over 300 AI-related laws and regulations in development globally. Major international processes and national regulatory approaches align closely on key areas such as fundamental rights, fairness, privacy, safety, transparency, competition, and accountability. However, there is less alignment in how different jurisdictions regulate, suggesting that international firms will need to manage divergence.

Executive Summary

AI's potential is vast, but so are its risks. As AI technologies evolve, governments and regulators are embarking on new regulatory activities. The report identifies nine broad areas of AI risk, including technical and security challenges, impacts on social dynamics, and long-term existential risks. Regulatory approaches are developing in a multilayered way, requiring organizations to have strong regulatory horizon scanning capabilities.

Chapter 1: From Fiction to Reality – Defining AI, Its Evolution, Opportunity, and Impact

AI is a pioneering technology with the potential to redefine human identity and societal functions. However, it also presents significant risks, including ethical concerns, technical challenges, and economic impacts. Understanding these risks is crucial for organizations developing or deploying AI.

Chapter 2: Key International AI Governance Processes and Risks

AI is a global issue, and international coordination is crucial. Key bodies like the UN, G7, and OECD are establishing shared AI principles and guidelines. The focus is on AI safety, with potential future binding international rules.

Chapter 3: Common Risks and Issues Across National and Regional Regulatory Approaches

National and regional approaches show strong convergence on key issues like human rights, fairness, and safety. However, there is divergence in execution and application, suggesting firms will face regulatory divergence.

Chapter 4: Key Themes Across All Regulatory Approaches

There is a high degree of convergence on key themes across international and national approaches. However, regulatory focus is concentrated on a subset of AI risks, and firms may need to manage broader risks alongside regulatory compliance.

Chapter 5: National and Regional Regulatory Approaches

There is divergence in how different jurisdictions approach AI regulation. Three broad approaches are identified: horizontal regulation, vertical regulation, and codes of conduct. Understanding the combined regulatory burden will be a key challenge for firms.

Chapter 6: How Can Companies Prepare for Regulatory Compliance?

Firms should focus on understanding the impact of regulations on their AI strategy, creating organizational clarity, engaging stakeholders, designing an AI governance framework, and implementing no-regret actions. Key actions include forming an AI governance committee, creating an AI system inventory, and conducting risk assessments.

Conclusion

AI regulation presents both challenges and opportunities. By understanding the regulatory landscape and implementing a comprehensive compliance strategy, firms can navigate the complexities of AI regulation while enabling innovation.

Contacts

Lead Author: Scott Bailey, Director, scottbailey@deloitte.co.uk

Key Deloitte France contacts: Richard Eudes, Managing Director, reudes@deloitte.fr; Nicolas Fleuret, Associé Responsable Risk & Resilience, nfleuret@deloitte.fr

Key contributors: Haarika Kanuparthy, Mark Hutcheon, Michael Greco, Valeria Gallo