A U.S. Department of Homeland Security seal is seen after a press conference near an international bridge connecting Mexico and the United States as U.S. authorities accelerate the removal of migrants at the Mexican border, September 19, 2021 in Del Rio, Texas.Reuters/Marco Bello/File Photo Obtaining license rights
WASHINGTON, Sept 14 (Reuters) – The Department of Homeland Security on Thursday highlighted the department’s successes in using artificial intelligence technology to help with drug busts and arrests of criminals, while also announcing new developments in the use of artificial intelligence. We plan to announce additional restrictions.
“We … ensure that the use of AI is responsible and trustworthy, that its effectiveness is rigorously tested, and that it protects privacy, civil rights and civil rights while avoiding inappropriate bias. We need to make sure that we are protecting freedoms and that … AI is transparent and accountable to the people we serve,” DHS Secretary Alejandro Mayorkas said today. This is stated in the AI policy memo to be released later.
The new policy comes as the agency is rapidly deploying artificial intelligence technology for a variety of sensitive missions, from border control to tracking the flow of fentanyl into the country.
DHS also hopes to use AI to improve its ability to secure U.S. supply chains and digital forensics capabilities in the future, despite the technology’s unique challenges, according to a senior official.
“I think the potential for unintended harm from the use of AI exists with any federal agency or use of AI,” said Eric Heisen, DHS’s chief information officer. “We interact with more people on a daily basis than any other federal agency, and our interactions with people are sometimes at the most critical times in their lives. .”
Historically, scholars have pointed to the dangers of AI when it comes to racial profiling, as AI can still make mistakes when identifying relationships in complex data.
As part of the new policy, Americans will be able to opt out of facial recognition technology in a variety of situations, including when checking in for flights.
According to the new directive, which the agency plans to release later Thursday with an AI memo, the guidelines require that facial recognition matches discovered using AI technology be manually performed by human analysts to ensure accuracy. It also requires a review.
Mr. Hisen is scheduled to address the recent incident at the San Ysidro, California, port of entry during a Congressional hearing on Thursday. In this incident, Customs and Border Protection agents used advanced machine learning (ML) models to issue warnings to unremarkable vehicles traveling north from Mexico. “Potentially suspicious pattern.” Investigators later found 75 kilograms of drugs in the car’s gas tank and rear quarter panel.
Another area where DHS is already making extensive use of AI technology is the southern border, where the agency has deployed more than 200 surveillance cameras, Heisen said.
Heisen said the cameras, sold by Anduril, a military defense contractor, use AI to automatically detect and flag areas people cross to thwart human and drug trafficking activities. It is said to be useful.
Reporting by Alexandra Alper and Christopher Bing.Editing: Chizu Nomiyama
Our standards: Thomson Reuters Trust Principles.
