The Centre for Long-Term Resilience (CLTR) has urged the UK government to establish a centralised system for recording incidents of artificial intelligence (AI) misuse and malfunctions.
The thinktank warns that without such a system, the government risks being unaware of critical issues involving AI technology.
Key Takeaways
- CLTR recommends a centralised system for logging AI incidents.
- Over 10,000 AI safety incidents have been recorded since 2014.
- The system would help the government respond quickly to AI-related issues.
- Incident reporting is already used in safety-critical industries like aviation and medicine.
- The UK government is urged to follow a similar approach for AI.
The Need for a Centralised System
The CLTR's report highlights the necessity of a centralised system for logging AI incidents, similar to the Air Accidents Investigation Branch (AAIB) in aviation. The thinktank argues that such a system is vital for the successful use of AI technology in public services and beyond.
The Organisation for Economic Co-operation and Development (OECD) has recorded 10,000 AI safety incidents since 2014. These incidents range from physical harm to economic, reputational, and psychological damages. Examples include deepfake videos, biased AI models, and self-driving car accidents.
Current Gaps in AI Regulation
The CLTR points out that the UK's current AI regulation is fragmented and lacks an effective incident reporting framework. This gap leaves the Department for Science, Innovation, and Technology (DSIT) without the necessary visibility to act swiftly on AI-related issues.
The thinktank emphasises that many AI incidents may not be covered by existing UK watchdogs, as there is no regulator focused on cutting-edge AI systems like chatbots and image generators. Labour has pledged to introduce binding regulations for the most advanced AI companies.
Benefits of Incident Reporting
Incident reporting has proven effective in other safety-critical industries. The CLTR believes that a similar approach for AI would provide quick insights into how AI systems are failing and help the government anticipate future incidents.
The system would also enable coordinated responses to serious incidents, where speed is crucial. Additionally, it would help identify early signs of large-scale harms that could occur in the future.
Recommendations for the UK Government
The CLTR has outlined three immediate steps for the UK government:
- Create a Government System for Reporting AI Incidents: This system should focus on public services and build on the existing algorithmic transparency reporting standard.
- Identify Gaps in AI Incident Reporting: Collaborate with UK regulators to find critical gaps in AI oversight.
- Pilot AI Incident Database: Develop a pilot database to collect AI-related episodes from existing bodies like the AAIB and the Information Commissioner's Office.
Conclusion
The CLTR's call for a centralised AI incident reporting system aims to transform the UK's approach to AI regulation. By adopting a proven method from other safety-critical industries, the UK can better manage the risks associated with AI technology and ensure a safer, more resilient future.
Sources
- UK needs system for recording AI misuse and malfunctions, thinktank says | Artificial intelligence (AI) | The Guardian, The Guardian.
- UK needs system to record AI misuse and mistakes, says thinktank, Computing UK.
- CLTR calls for central UK system to record AI malfunctions, misuse - Telecompaper, Telecompaper.
- UK thinktank urges AI incident reporting system for Government - Techerati, Techerati.
