Homeland Security Adopts and Advances Artificial Intelligence
January 17, 2024
Artificial Intelligence (AI) has rapidly become a familiar term. While most Americans are only vaguely aware of what this transformative technology can accomplish, federal departments and agencies are taking significant steps to harness the power of AI to advance their missions, improve the efficiency and efficacy of their operations, and establish safeguards against the “increasingly sophisticated threat” to America’s critical infrastructure by adversarial artificial intelligence (AAI).
The Department of Homeland Security (DHS) has been at the forefront of AI adoption at the federal level. The DHS mission is to “secure the nation from the many threats we face,” and the agency regards AI as both a tool to leverage against challenges and threats, as well as a potential source of the threats it protects America against.
On April 20, 2023, DHS Secretary Alejandro Mayorkas established the first DHS AI Taskforce (AITF). Led by Under Secretary for Science and Technology Dimitri Kusnezov and DHS Chief Information Officer Eric Hysen, the AITF advances AI applications across four priority initiatives critical to homeland security:
- Enhance the integrity of supply chains and broader trade.
- Better detect the flow of fentanyl into the United States and disrupt the criminal networks that traffic the deadly drug.
- Bolster digital forensic tools help identify, locate, and rescue victims of online child sexual exploitation and abuse, as well as identify and apprehend the perpetrators of these crimes.
- Assess AI’s impact on DHS’ ability to secure critical infrastructure by working with partners in government, industry, and academia.
AI Keeps Narcotics Out of America
In his September 2023 testimony before a House of Representatives Committee, AITF Co-Chair Eric Hysen explained how federal agencies were harnessing AI. Hysen described how AI technology recently led to the seizure of 75 kilograms of narcotics concealed within a vehicle attempting to enter the United States at the San Ysidro Port of Entry (POE) in San Diego, California. Historically, Customs and Border Patrol (CBP) Agents might not have had any cause to inspect this specific vehicle more rigorously than others. However, one of
CBP’s advanced AI machine learning (ML) tools identified within seconds that this vehicle had a suspicious border crossing pattern and alerted officers immediately. According to Hysen,
“If not for this use of AI, these narcotics could have been on our streets and those smugglers could have been admitted into our country.”
Inspection referrals from these ML models have led to 240 seizures at America’s borders, which included thousands of kilograms of narcotics. AI technology has helped CBP and DHS rigorously check each vehicle and person entering the United States while not disrupting America’s economy by drastically slowing down trade.
Hysen’s testimony included other examples of AI use to advance DHS missions, citing that AI tools will allow agents and officers to focus on investigations and make more informed decisions by decreasing the time they spend on manual data entry and analysis. For example, the:
- Federal Emergency Management Agency (FEMA) uses AI to simplify grant application processing.
- U.S. Coast Guard employs computer vision technology for maritime small object detection.
- Transportation Security Agency (TSA) uses ML tools to confirm travelers’ identities through digital identification cards on their mobile devices, reducing airport lines and wait times.
The DHS’ adoption of AI is only just beginning. In October 2023, DHS Undersecretary for Policy Robert Silvers attended an AI conference hosted by Critical Infrastructure Technology (ICIT) in Washington, D.C. After the conference, the Federal News Network interviewed Silver. He stated that DHS’ adoption of AI should be “early and aggressive” and that AI technologies, when appropriately used, can significantly bolster DHS’ ability to meet its security mission. Silvers expressed that DHS is pushing forward with its adoption of AI but acknowledged that offensive actors will also use this technology for adversarial purposes. He stated that catastrophic cyber-attacks using AI have not yet occurred, and it is unclear whether the cyber defenders or malicious actors will attain AI’s upper hand.
Despite the seemingly fast-paced embrace of AI at DHS, the department has implemented measures to ensure that these new technologies protect and uphold civil liberties, civil rights, and privacy to ensure that DHS’s adoption of AI retains the public’s trust. On September 14, 2023, DHS announced new measures to promote the responsible use of AI, including a directive on using facial recognition and capture technologies to safeguard against unintended bias or discrimination. In addition, citizens can opt out of facial recognition for non-law enforcement uses.
“Deployments in new technology over the last 10 years have dramatically increased our ability to interdict narcotics and weapons, disrupt transnational criminal and human smuggling operations, and rescue countless individuals victimized by unscrupulous smugglers,” DHS Spokesperson Marsha Espinosa told the Washington Post.
“DHS will continue to invest in technologies that increase its operational advantage at our borders.”
Leveraging AI requires both short- and long-term strategies to enhance DHS’s ability to protect America while transforming the operational and organizational landscape of the Department.
How WBD is Helping
Here at WBD, we recognize the benefits of AI and other advanced data-analytic capabilities to support our federal government clients. In one example, our acquisition specialists and data scientists collaborated to develop a cutting-edge machine-learning tool that streamlines workflows, improves process transparency, and manages complex document flows. The WBD-designed Acquisition Requirement Workflow, or Arrow tool, managed a $1.2 billion contract portfolio and more than 800 annual contract actions annually for our DoD clients.
“Our government operates on Microsoft platforms, and the DoD specifically has a lot of security considerations already woven into SharePoint. So, we designed a tool using the existing system architecture to keep files secure,” says Nicolas Rivard, Vice President of WBD, who helped develop Arrow.
Arrow assisted government contracting officers, offices, and requirement owners in developing and processing acquisition packages for an award. Users submitted their requirements for an award into a simple-to-use input form with the capability to upload supporting documents. After intake, the tool assisted users and managers in tracking the development of the package and allow for monitoring of all inputs, outputs, discussions, and decisions.
“They didn’t need to trust or load any new software. The Arrow tool we designed made advanced use of SharePoint’s existing capabilities,” says Rivard.
Author: Benjamin Hein is a Senior Associate at WBD supporting the FEMA Office of the Chief Information Officer.