20 février 2018 | International, Aérospatial, Naval, Terrestre, C4ISR, Sécurité

AI makes Mattis question ‘fundamental’ beliefs about war

By: Aaron Mehta

WASHINGTON – Over the years, Secretary of Defense Jim Mattis has cultivated a reputation for deep thinking about the nature of warfare. And during that time, he has come to a few conclusions about what he calls the “fundamental” nature of combat.

“It's equipment, technology, courage, competence, integration of capabilities, fear, cowardice — all these things mixed together into a very fundamentally unpredictable fundamental nature of war,” Mattis explained Feb. 17. “The fundamental nature of war is almost like H20, ok? You know what it is.”

Except, that might not be true anymore.

During a return flight from Europe, Mattis was asked about artificial intelligence — a national priority for industry and defense departments across the globe, and one driving major investments within the Pentagon — and what the long-term impact of intelligent machines on the nature of war might be.

“I'm certainly questioning my original premise that the fundamental nature of war will not change. You've got to question that now. I just don't have the answers yet,” he said.

It's both a big-picture, heady question, and one that the department needs to get its head around in the coming years as it looks to offload more and more requirements onto AI. And it's a different question than the undeniable changes that will be coming to what Mattis differentiated as the character, not nature, of war.

“The character of war changes all the time. An old dead German [Carl von Clausewitz] called it a ‘chameleon.' He said it changes to adapt to its time, to the technology, to the terrain, all these things,” Mattis said.

He also noted that the Defense Innovation Board, a group of Silicon Valley experts who were formed by previous defense secretary Ash Carter, has been advising him specifically on AI issues.

For now, the Pentagon is focused on man-machine teaming, emphasizing how AI can help pilots and operators make better decisions. But should the technology develop the way it is expected to, removing a man from the loop could allow machine warfare to be fully unleashed. Mattis and his successors will have to grapple with the question of whether AI so radically changes everything, that war itself may not resemble what it has been for the entirety of human history.

Or as Mattis put it, “If we ever get to the point where it's completely on automatic pilot and we're all spectators, then it's no longer serving a political purpose. And conflict is a social problem that needs social solutions.”

https://www.defensenews.com/intel-geoint/2018/02/17/ai-makes-mattis-question-fundamental-beliefs-about-war/

Sur le même sujet

  • Here’s why Britain is struggling to form a fully effective carrier strike group

    29 juin 2020 | International, Naval

    Here’s why Britain is struggling to form a fully effective carrier strike group

    By: Andrew Chuter LONDON — Britain's Royal Navy took delivery of two new aircraft carriers, but a government report on the ships achieving operational capability has laid bare some obstacles toward making a fully effective carrier strike group. In a report released June 25, the National Audit Office pointed to delays in developing the Crowsnest airborne early warning radar and contracting to build the logistics ships destined to support the 65,000-ton carriers as ongoing problems for the Royal Navy. The NAO also raised questions about future funding. The Ministry of Defence is making slow “progress in developing the crucial supporting activities that are needed to make full use of a carrier strike group, such as the Crowsnest radar system and the ability to resupply the carriers. In addition, it has not established a clear view on the future cost of enhancing, operating and supporting carrier strike, which creates the risk of future affordability pressures,” the NAO said. Added the head of the watchdog: “The MoD also needs to get a firmer grip on the future costs of carrier strike. By failing to understand their full extent, it risks adding to the financial strain on a defense budget that is already unaffordable.” HMS Queen Elizabeth, the first of two carriers built by British industry in a £4.6 billion (U.S. $5.7 billion) program, is already undertaking extensive sea trials, with its F-35B jets ahead of a planned first deployment next year. The second carrier, HMS Prince of Wales, is also conducting sea trials but is some months behind its sister ship. The warships are not expected to be operated simultaneously. The NAO said the Lockheed Martin-led program to install Crowsnest radars on Royal Navy Merlin helicopters is running 18 months late and will impact how the British carrier strike force is initially deployed. The watchdog said the MoD is working to come up with an acceptable baseline radar by the time HMS Queen Elizabeth undertakes its initial deployment next year. “As at April 2020, the Department [the MoD] expected to achieve initial Crowsnest operating capability in September 2021, some 18 months later than planned,” the NAO reported. “As this is later than the December 2020 milestone for declaring initial operating capability for carrier strike, the Department is working to provide a credible baseline radar capability for the first deployment with the United States in 2021. It expects to recover some lost time to declare full operating capability in May 2023, 11 months later than planned. However, the existing timetable contains no contingency to accommodate any further slippage. The delays will affect how the Department can use carrier strike during this period.” British and U.S. Marine Corps jets will be based on the carrier during its first deployment, partly because the U.K. does not have a sufficient inventory of available jets. Eighteen of the aircraft have so far been delivered for use by the Royal Navy and the Royal Air Force Lockheed Martin secured the Crowsnest contract in 2017, with Searchwater radar supplier Thales and helicopter builder Leonardo as subcontractors. Crowsnest is a key element in the protection of the naval strike group, giving air, maritime and land detection and tracking capabilities. The NAO said the delay “has been caused by a subcontractor, Thales, failing to meet its contractual commitments for developing equipment and not providing sufficient information on the project's progress. Neither MoD nor its prime contractor, Lockheed Martin, were aware of these problems until it was too late, reflecting MoD's ineffective oversight of its contract with Lockheed Martin.” A Lockheed Martin UK spokesperson said the company is working to deliver the Crowsnest capability in time for HMS Queen Elizabeth's deployment. “As prime contractor for Crowsnest, we understand the fundamental component that this program delivers to the UK's Carrier Enabled Power Projection. We will continue to ensure that the program develops in line with our requirement to deliver the Crowsnest capability to support the first operational deployment of HMS Queen Elizabeth,” the spokesperson said. “We will work with our industrial partners and the MoD to address any developmental issues which arise, including the deployment of additional resources, if necessary, to maintain program timescales and deliver this critical capability to the Royal Navy.” Thales UK did not respond to Defense News' requests for comment by press time. The NAO partly blames the setbacks for why the MoD faces a “tight timetable” to develop full operating capability for a strike group by 2023. But the watchdog also highlighted the Fleet Solid Support program as another obstacle. The MoD had targeted 2026 for when the first of up to three logistics ships could provide ammunition, food and general stores to the carrier strike group, but that timeline has extended by up to three years as a result of ongoing uncertainty over the schedule to compete and build the vessels operated by the Royal Fleet Auxiliary. The MoD abandoned a competition to build the ships late last year, saying it was concerned about obtaining value for money. At the time, the program was mired in controversy over whether the contract should go to a British shipyard consortium or awarded to a foreign company. That issue remains unresolved. No date has officially been given for restarting the competition. Defence Secretary Ben Wallace told the parliamentary Defence Committee earlier this year that he thinks it will relaunch in September, but that has not been confirmed. Defence Committee Chairman Tobia Ellwood was particularly critical of the failure to provide the necessary support ships, noting that without them, the carriers' capability would be seriously undermined. “It'll be hotched and potched, only available for short operational journeys,” he told the Daily Telegraph on June 26. “It will be for display purposes only, and that's a very expensive toy.” Britain has only one solid support vessel, RFA Fort Victoria, that can replenish a carrier at sea. It entered service in 1994 and is due to retire in 2028, having had its life expectancy extended. The NAO report said the limitations of RFA Fort Victoria would have a knock-on effect to carrier operations. “Having only one support ship with limited cargo capacity slows the tempo and reach at which the Department [the MoD] can replenish a carrier group. In addition, the Department will have restricted options for deploying the carriers for much of 2022 because RFA Fort Victoria will be unavailable due to major planned maintenance work,” the NAO said. Responding to the report, an MOD spokesperson said: ”Carrier strike is a complex challenge, which relies on a mix of capabilities and platforms. We remain committed to investing in this capability, which demonstrates the U.K.'s global role. “Despite the disruptions of COVID-19, the carrier strike group is on track for its first operational deployment.” https://www.defensenews.com/naval/2020/06/26/heres-why-britain-is-struggling-to-form-a-fully-effective-carrier-strike-group/

  • NGA taking a 'try before you buy' approach to commercial solutions

    7 octobre 2021 | International, Aérospatial

    NGA taking a 'try before you buy' approach to commercial solutions

    The agency is using bailment agreements and other contracting tools to test out commercial services before committing to them, allowing it to evaluate and integrate those services faster.

  • Contracts for November 3, 2021

    4 novembre 2021 | International, Aérospatial, Naval, Terrestre, C4ISR, Sécurité

    Contracts for November 3, 2021

    Today

Toutes les nouvelles