NATO has released a revised AI technique to promote the responsible use of AI in defense applications and to combat threats from AI-enabled adversaries.
NATO’s updated strategy is a sign of how quickly AI in defense is moving from novelty to broad adoption.
NATO’s original AI strategy was set out in 2021 where it endorsed six Principles of Responsible Use (PRUs) for AI in defense, namely: Responsibility and Accountability, Explainability and Traceability, Reliability, Governability and Bias Mitigation.
2021 was only 3 years ago but it surely’s several generations behind the commercially available tech we’ve got today. NATO says that as recent capabilities emerge and AI becomes a general-purpose technology, recent risks have emerged together with it.
While NATO is mostly related to national defense, the strategy says that other AI-related issues warrant NATO’s attention. The strategy noted “the potential diminishing global availability of quality public data to coach AI models.”
It also mentioned the implications of compute-intensive AI on energy consumption. NATO can also be concerned about “accountability in human-machine teaming and overcoming technical and governance issues when civilian-market dual-use solutions are applied in a military context.”
The desired outcomes expressed within the strategy make it clear that the weapons NATO members might be using will incorporate AI.
NATO says it’ll take “Measurable steps to integrate AI, enabled by quality data, into appropriate Allied capabilities through commitments within the NATO Defence Planning Process.”
To achieve its objective, NATO says it and its allies “have to find a way to access and use specialised laboratories, sandboxes and testing facilities.”
Unlike the Manhattan Project, cutting-edge AI development is going on in corporate labs, not government facilities. You’ve got to wonder if OpenAI or Meta would share their facilities if NATO asked them to.
NATO’s strategy also notes the impact that AI may have on military and civilian jobs. To take care of this can require “retraining programs, high level expertise, changes in job roles and integrating technical experts more deeply into military operations.”
Autonomous jets, robot soldiers, and swarms of drones will likely leave 1000’s of soldiers on the lookout for alternate employment.
It’s easy to default to the concept that wars are fought with military hardware, but NATO’s strategy incorporates this interesting insight into what its leaders are especially concerned about:
“Disinformation, the weaponization of gendered narratives, technologically facilitated gender-based violence and AI-enabled information operations might affect the end result of elections, sow division and confusion across the Alliance, demobilize and demoralize societies and militaries in times of conflict in addition to lower trust in institutions and authorities of importance to the Alliance. These issues could raise profound implications for the Alliance.”
In a scenario where nation-states use superintelligent AIs to compete with one another, strategic decisions, creation and distribution of propaganda, and outmaneuvering will occur in milliseconds.
The challenge that NATO members face is that essentially the most powerful AI models belong to corporations that wish to sell them into a world market.
NATO says it needs to search out a strategy to “mitigate the danger of Allied technology being exploited by potential adversaries and strategic competitors, and help Allies to safeguard access to vital components.”
Considering the advancements China has made despite US sanctions, that AI horse may have already got bolted.
