Concerns about AI harm lead to calls for UK incident reporting system

The Centre for Long-Term Resilience (CLTR), a UK think tank focused on long-term crises and extreme risks, is calling on the government to introduce a new reporting system for AI incidents.

Describing a “worrying gap in the UK’s regulatory plans”, CLTR said the system should monitor the extent to which AI creates safety risks in the real world and how it is regulated and used.

It should coordinate the response to major incidents where speed is of the essence, conduct root cause investigations and identify early warning signs of major damage that could occur in the future.

This information could then be used to assist the AI Safety Institute and the Central AI Risk Function in conducting risk assessments.

“(The Department of Science, Innovation and Technology) DSIT lacks a central, up-to-date picture of these types of incidents when they occur,” the report warned.

“While some regulators collect some incident reports, we believe this is unlikely to capture the novel harms caused by disruptive AI. DSIT’s priority should be to ensure that the UK government learns of such novel harms not through the news but through established incident reporting processes.”

Without such a system, the DSIT could fail to address fundamental model problems, such as bias and discrimination or misguided actors, incidents related to the use of AI in public services by the government itself, or misuse of AI systems, such as for disinformation campaigns or even the development of biological weapons.

To get things moving, the government should start with simple steps, such as expanding the Algorithmic Transparency Recording Standard (ATRS) to include a framework for reporting AI incidents in the public sector. These incidents could be reported directly to a government agency and potentially shared with the public.

It should engage UK regulators and consult experts to identify the most concerning vulnerabilities and ensure that incidents are treated with high priority.

In addition, the think tank said the government should also build capacity within the DSIT to monitor, investigate and respond to incidents, possibly by setting up a pilot database of AI incidents.

“This could form part of the ‘central function’ of the DSIT and begin to develop the policy and technical infrastructure for collecting and responding to AI incident reports,” the authors say.

Veera Siivonen, CCO and partner at AI governance firm Saidot, agreed with the recommendations, but cautioned that while implementation is an important first step, there is still a long way to go.

“As the UK heads towards the general election, the next government’s AI policy will be the cornerstone for economic growth. But this will require precision in finding a balance between regulation and innovation, by creating guardrails without limiting the industry’s potential for experimentation,” said Siivonen.

“The new UK government should provide certainty and understanding to businesses through clear governance requirements, while monitoring and mitigating the most likely risks,” she added.

“By integrating a variety of AI governance strategies with centralised incident reporting, the UK can harness the economic potential of AI and ensure it benefits society while protecting democratic processes and public trust.”

Related Posts

William Byron on Ryan Blaney incident: ‘I hate that this happened’

New Hampshire woman arrested after accident at Bedford toll booth

Shooting at 16th Street and Mission leaves one dead