Homeland Security boss says America's war on drugs and crime will be powered by AI.

WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

RSAC U.S. Homeland Security Secretary Alejandro Mayorkas says AI is a double-edged sword in that the government can see ways in which the technology can be used to protect and also be used to attack Americans. Is.

On the one hand, artificial intelligence can automate computer networks to defend against threats to America's critical infrastructure and protect the nation's citizens from harm.

However, terrorists and criminals are also apparently using the technology to launch automated attacks against these same critical assets and perpetuate crimes including child sexual exploitation and abuse — some of the things that Homeland Security is working against. .

Whenever government officials talk about the use of AI for everyday tasks, the usual concerns come to mind, such as the misuse of surveillance technologies by agents as well as the biases inherent in machine learning.

Addressing those concerns, Meyerkas said his department's Office of Civil Rights and Civil Liberties is at the “institutionalization” level to balance the interests of protecting the land of the free with the need to respect citizens' privacy. I can create a balance.

Homeland Security also created its Artificial Intelligence Safety and Security Board last month to address the issue. And in February, it launched the AI ​​Corps initiative, which aims to hire and connect 50 technologists in the field this year.

Meyerkas said the AI ​​Safety and Security Board met for the first time this week. And despite criticism that the board is stacked primarily with big tech giants — who may be inclined to put profits ahead of people's privacy and safety — responsible AI advocate Roman Chaudhary, who also sits on the safety board. are, and Meyerkas pushes back against this criticism. Their joint keynote.

“What we heard yesterday is an expression of the fact that the civil liberties, the civil rights implications of AI are really part and parcel of security,” Meyerkas said. “For example, we cannot consider the safe and secure use of AI as a potential continuation of implicit bias.”

He also pointed to three pilot programs underway at Homeland Security that Meyerkas said help the department advance its mission. One involves using LLMs to assist with homeland security investigations.

“We might have a task force investigating a drug case on the West Coast, and a different group of agents working on an international money laundering scheme on the East Coast,” Meyerkas explained.

Now we're entering their criminal investigation reports into a database, and we'll be able to use AI to identify connections.

“But now we're entering their criminal investigation reports into a database, and we'll be able to use AI to identify connections we wouldn't otherwise know about,” he added.

Meyerkas said another pilot will help the Federal Emergency Management Agency (known as FEMA) to “apply for under-resourced, target-rich communities for grants to ensure they have the right to vote.” Don't miss out on yogurt,” Meyerkas said.

These are funds released by the federal government that are available for emergency and post-emergency and disaster relief assistance to communities affected by natural disasters and other major events.

And a third program will use LLMs to train Homeland Security officers who work with refugees and asylum seekers who apply for citizenship in the United States.

“An asylum officer in training can now actually ask questions to a machine that has been trained to act as an asylum seeker, both in terms and style,” Meyerkas said.

By “stylistically,” he explained: “Often, people who have been through trauma are not forthcoming in revealing traumatic experiences, and we've trained the machine to be similarly gentle. ” ®

WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

Leave a Comment