AI in Federal Law Enforcement

AI in Federal Law Enforcement: Enhancing Operational Efficiency

Federal law enforcement agencies are increasingly turning to AI to streamline operations and improve public safety. At the recent AFCEA Bethesda’s LEAPS Preview event, experts from the FBI, TSA, and NCIS discussed the strategic integration of AI technologies, including machine learning, automation, and computer vision, into their workflows. While these agencies are proceeding cautiously due to legal and ethical considerations, they are already seeing significant benefits in areas such as data processing, suspect identification, and operational efficiency.

Moderated by Sonya Thompson, former chief information officer at the Federal Bureau of Prisons, the event highlighted how AI is transforming federal law enforcement. The FBI, TSA, and NCIS are exploring a range of AI applications, from computer vision tools to facial recognition and AI-driven training programs. These technologies are helping agencies process large datasets more quickly, enhance investigative capabilities, and improve decision-making.

FBI's AI Innovations

The FBI’s adoption of AI began in earnest after the 2013 Boston Marathon bombing, when the agency needed to quickly analyze vast amounts of video data to locate suspects. According to Kiersten Schiliro, senior technical advisor at the FBI, the agency developed an open Multimedia Processing Framework (MPF) to extract specific optical characteristics from video footage. This tool can identify license plates, objects like vehicles, and even track faces across datasets, though it does not perform identity resolution.

The MPF has significantly reduced the time required to review data, a critical advantage given the increasing use of body-worn cameras and other video sources. Schiliro noted that processing the Boston Marathon bombing data took nearly a year at the time but would now take only two days with current AI capabilities. The FBI is also using AI in its Criminal Investigative Division to combat violence against children, employing facial recognition technology to identify vulnerable victims who may not appear in adult databases.

Schiliro emphasized that the FBI is strategically applying AI to its most challenging problem sets first, which has made adoption easier within specific divisions. The bureau is exploring additional AI use cases to increase accuracy, efficiency, and capabilities beyond human limitations. Schiliro advised AI providers to focus on measurable outcomes for each AI application, ensuring that the technology delivers tangible benefits.

TSA's AI Integration

The Transportation Security Administration (TSA) is leveraging AI tools developed by its parent organization, the Department of Homeland Security (DHS). Kristin Ruiz, deputy assistant administrator and deputy chief information officer at the TSA, explained that the agency is training its IT staff to implement AI solutions without needing additional personnel. The TSA is using tools like the TSA Answer Engine, which allows employees to quickly access standard operating procedures, generate reports, and get answers to regulatory questions.

The TSA’s Innovation Lab is also playing a key role in AI implementation. The lab partners with industry experts to explore use cases and demonstrate how AI can automate tasks. For example, the TSA recently used virtual reality holograms combined with ChatGPT personas to simulate training scenarios at airport checkpoints. This approach allowed officers to practice handling different personas and situations in a controlled environment.

Ruiz highlighted the TSA’s collaboration with other agencies, including the DHS, Department of Justice, and State Department. By sharing data and leveraging AI for facial recognition and biometric analysis, the TSA is enhancing security at airports and ports while improving operational efficiency.

NCIS' AI Initiatives

The Naval Criminal Investigative Service (NCIS) is also exploring AI applications, starting with small projects to enhance operational capabilities. Richard Dunwoodie, acting executive assistant director in the NCIS’ Operational Technology and Cyber Innovation Directorate, noted that AI offers significant potential for the agency, both in business applications and operational environments. The NCIS is focusing on AI tools for vehicle recognition and human identification, as well as efficiency gains in policy navigation.

Dunwoodie emphasized the importance of vetting and validating AI solutions to ensure they meet the agency’s security and operational standards. The NCIS is working closely with the Department of Defense’s Chief Digital and Artificial Intelligence Office (CDAO) to identify approved AI technologies that can be quickly integrated into their workflows. Dunwoodie also stressed the need for staff education and policies to prevent the misuse of commercial AI tools, which could potentially expose sensitive information to adversaries.

Schiliro added that the FBI has established robust checks and balances for AI use, including privacy and civil liberties reviews and ethics assessments by the FBI’s AI Ethics Council. This cautious approach ensures that AI is implemented responsibly and effectively across federal law enforcement agencies.

The Future of AI in Federal Law Enforcement

As AI technologies continue to advance, federal law enforcement agencies are poised to benefit even further. AI has the potential to revolutionize investigative processes, enhance public safety, and streamline operations. However, agencies must navigate legal, ethical, and technical challenges to ensure that AI is used responsibly and effectively. By adopting a measured approach and focusing on measurable outcomes, federal law enforcement can harness the power of AI to meet the evolving demands of public safety.