The Home Office is beginning to sound out the technology industry and academia on the threats and opportunities for policing in the emergence of AI.
Its National Policing Capabilities Unit (NPCU) is running a scoping exercise with IT industry association techUK on the implications of advances in AI for public safety.
It will take into account the relevant legislation, regulations, guidelines for the use of AI and funding for its development, testing and roll out.
The exercise will involve a series of round tables and visits, and complements work being led by the policing chief scientific advisor.
techUK has invited suppliers to provide input into the exercise through an online form. It includes questions on what capabilities the police might need to respond to AI related threats, how safeguards can be built into products to help crime prevention, and how to overcome some of the existing challenges in digital policing.
Uncertainties
In 2021 a UK Parliament briefing paper said there are uncertainties around the extent and impact of AI in policing and security.
It highlighted potential benefits including improving public safety and helping police forces and security services deal with limited resources; but it also pointed to concerns such as the effects of bias within algorithms, particularly in facial recognition, and the risk of directing police officers towards areas that are already over-policed.
The paper raised a series of questions such as how the relevant governance can be improved, what guidance is needed for the detection and mitigation of bias, and should any systems used be subject to a mandatory audit or certification process?
