In a series of webinars over the early summer we heard from a wide range of industry experts on the challenges and opportunities that AI can bring the Justice & Police Sectors.

In this final video from the first webinar we hear from all contributors asking questions from over 500 audience members:

• Paul Taylor, The Police Chief Scientific Advisor for the National Police Chiefs’ Council
• Dr Eric Halford, former Detective Chief Inspector and now Assistant Professor, Policing and Security Program from the Randam Academy
• Dr Dory Reiling , retired Senior Judge, Lead reporter European Law , Institute for Digitalisation , Section Editor International Law Journal ( Curt Administration )
• Michael O’Connell Managing Director ,Critical Insights Consultany and a Senior fellow at the United Nations Interregional Crime and Justice Research Institute

The questions covered various topics related to the use of AI in law enforcement. Some of the key themes discussed include the lack of a nationally agreed policing model for AI, the importance of understanding the difference between AI and machine learning, the need for transparency and education around AI, the potential use of AI in reviewing chat logs and open source intelligence, the challenges of using AI in multi-agency collaborations, the potential for AI to assist in building case files and taking statements, the need for ethical oversight and regulation of AI in policing, the risks of AI being compromised by criminal actors, and the challenges of auditing and ensuring the reliability of AI-generated outputs.

Takeaways

• There is currently no nationally agreed policing model for AI, highlighting the need for consistency in how AI is implemented in law enforcement.
• It is important to differentiate between AI and machine learning and to use informed language when discussing AI.
• Transparency and education are crucial in ensuring the responsible and effective use of AI in law enforcement.
• AI has the potential to assist in reviewing chat logs and open source intelligence, but there are challenges in ensuring accuracy and reliability.
• Multi-agency collaborations can benefit from the use of AI, but careful consideration must be given to data sharing and interoperability.
• AI can be used to enhance efficiency in building case files and taking statements, but proper policies and processes must be in place to ensure accuracy and accountability.
• Ethical oversight and regulation are necessary to address the potential risks and biases associated with AI in policing.
• The use of AI in law enforcement may be compromised by criminal actors, highlighting the need for robust cybersecurity measures.
• Auditing and ensuring the reliability of AI-generated outputs is a challenge that requires careful data management and accountability.

Sound Bites

“Transparency and education are crucial in ensuring the responsible and effective use of AI in law enforcement.”
“There needs to be procedural justice in the decisions that are actually taken once unwanted content has been detected.”
“Judiciaries themselves need to start thinking about how they will frame their own use of AI.”

Chapters

00:00 Introduction and Lack of a Nationally Agreed Policing Model for AI

02:06 Differentiating AI and Machine Learning

03:01 Transparency and Education in AI

06:36 AI in Reviewing Chat Logs and Open Source Intelligence

09:42 Challenges of Multi-Agency Collaborations with AI

13:14 Enhancing Efficiency in Building Case Files and Taking Statements with AI

26:48 Addressing the Risks of AI Compromise by Criminal Actors

33:13 Auditing and Ensuring the Reliability of AI-Generated Outputs  

We have launched a YouTube channel where you can also view our videos – just click here and please like and subscribe PTV Youtube