Free beta access. 50 free AI tokens each week. Optional top-ups available. See what is included
Opening Word document…

3. TURN: AI increases nations vulnerability to cyber-attacks.

James Johnson, 2020 (Postdoctoral Research Fellow at the James Martin Center for Nonprolifera¬tion Studies, WASHINGTON QUARTERLY, “Artificial Intelligence in Nuclear Warfare: A Perfect Storm of Instability?”)

It is now thought to be possible that a cyberattack could infiltrate a nuclear weapons system, threaten the integrity of its communications, and ultimately (possibly unbeknownst to its target) gain control of its nuclear as well as non¬nuclear command and control systems. Advances in AI might, therefore, exacer¬bate this challenge by enabling improvements to the cyber offense, thereby con¬ferring further advantages to first movers in this domain.14 For example, machine learning and AI could dramatically reduce the high levels of labor inten¬sity and technical skill required to execute advanced persistent threat (APT) oper¬ations—or “hunting for weaknesses.”15 Future AI APT tools could offer a would-be attacker a cheap and easy way to replicate powerful cyber weapons, which would require minimal technical knowledge to employ. While manipulations and system subversions are possible with existing cyber offense tools, AI and increasing degrees of military autonomy could amplify the potential speed, power, and scale of future attacks in cyberspace. In response to these anticipated vulnerabilities and to achieve the first-mover advantage, China, Russia, and the United States have continued to harden their AI cyber defenses.16 For example, open sources indicate that Chinese analysts view the vul¬nerability of China’s nuclear command, control, and communication (NC3) systems to cyber infiltrations as highly escalatory, even if an attacker’s objective was limited to cyber espionage.17 China’s fear that its early-warning missile systems are inadequate to respond to a disarming first strike by the United States has prompted Beijing to prioritize mitigating false negatives, when early-warning systems fail to warn of an imminent attack (as opposed to a false positive that warns of an attack that does not exist).18 AI-enhanced cyber capabilities sim¬ultaneously bolstering deterrence and incentivizing others to attack could intensify a paradox of enhanced capabilities and increased vulnerabilities in the cyber domain.

Sign in to recut, update, or save this card.