Defence

The Pentagon is approaching the implementation of “ethical” artificial intelligence

The Department of Defense is still finalizing a plan to implement its ethical principles of artificial intelligence, said Jane Pinelis, head of the artificial intelligence department at the Joint Center for Artificial Intelligence of the Ministry of Defense, at an event Tuesday.

“Thus, we are the first military personnel to adopt ethical principles for AI. Since then several other countries have done so and where we are now with [chief digital and artificial intelligence office] “We are trying to move towards implementation,” Pinelis said during a panel discussion at the Atlantic Council on May 17.

“So we have five ethical principles at the moment [gotten] instructing the Deputy Secretary to promote them on six different principles. But now we are moving towards … implementation. ”

Department of Defense released instructions for the responsible use of AI in May 2021 after the announcement of the set ethical principles years earlier.

Pinelis said the implementation plan, pending the signature of the deputy defense minister, would be a “formal way forward” that tasked “various organizations in the defense ministry with very concrete actions to actually put these principles into practice”.

Many of these tasks, she continued, coincide with testing and evaluation, but there are many parts that require everyone in the U.S. Department of Defense to have some responsibility.

“Responsible artificial intelligence is like everyone’s work in the department,” Pinelis said. “And so there are parts of it that are relevant to international allies. There are parts of that that are relevant to the responsible acquisition of these systems and the responsible design of these systems, and, again, to the creation of all these arguments and evidence that go to responsible AI ».

Michael Horowitz, director of the Department of Defense’s New Opportunities Policy, said that faster support for artificial intelligence and autonomous technology solutions requires budget support and centralized leadership – both of which the Pentagon is working to address by establishing its head office for digital and artificial intelligence. .

“If data is the fuel that makes AI work in essence – what is an algorithm without data that you would use to train it one way or another – then combining them into [chief digital and artificial intelligence office] “I think it will be reflected in what the new strategy will look like,” Horowitz said during the main panel at the event. “What is needed now is to turn these thoughts into reality and make it faster.”

Horowitz, who has been in a brand new role for about a month, said he was “quite optimistic” about the Pentagon’s direction and focus on AI and autonomy thanks to the creation of the Office of New Opportunities Policy, CDAO and the Innovation Steering Group. that Deputy Secretary of Defense for Research and Technology.

“I think all these things make me optimistic that when we enter the fiscal cycle of the 24th fiscal year, we will start to see this payout as the department becomes – it’s not just more, but more it is wise to think about AI and autonomous systems and investments in such a way that the combined forces really pay off ”.



https://www.defenseone.com/policy/2022/05/pentagon-closing-ethical-ai-implementation/367120/ The Pentagon is approaching the implementation of “ethical” artificial intelligence

Back to top button