The Cybersecurity and Infrastructure Security Agency (CISA) plans to release a directive to federal agencies detailing actions required to carry out the president’s artificial intelligence executive order by the end of the week, CISA Acting Director Nick Andersen said Wednesday.

The binding operational directive will focus in part on “vulnerability alleviation and vulnerability management,” Andersen said in remarks delivered at the TechNet Cyber conference in Baltimore.

CISA also will be rolling out “specific artificial intelligence access” to partners in the coming days, Andersen said.

The artificial intelligence executive order released Tuesday is a scaled-back version of an earlier iteration that was spiked amid internal conflict within the administration and concerns raised by former artificial intelligence and crypto czar David Sacks.

The latest version of the order asks companies to voluntarily submit models to the government for testing 30 days before they are released publicly. Originally, the administration had asked for 90 days.

The potential risks posed by some models are important to consider, Andersen said, but he also focused on how AI can bolster cybersecurity protections.

“How can we actually use it as a good defensive tool and how is it going to help us reduce our attack surface exposure?” Andersen said. 

CISA will play a key role in helping to stand up the “cyber clearinghouse” the executive order envisions, Andersen said, and it also will be accessing models to vet. 

The government needs to do a great deal of work to address the ever increasing threat environment, he added.

“The larger problem we're having to address here is we kick the can down the road in a fairly significant way with our IT infrastructure,” he said. “We have end of life limited service devices that are operating within our environments… Our adversaries can reach in and touch us.”