A leading security think tank has warned that a devastating nuclear war that wipes out humanity could be brought about by artificial intelligence as soon as 2040.
It further claims that advances leading to ‘doomsday AI’ machines could encourage nations to take apocalyptic risks with their nuclear arsenals.
During the Cold War, a condition called MAD (mutually assure destruction) maintained an uneasy truce between superpowers as both sides had little incentive to launch a nuclear attack, as it would have been met by a devastating retaliation that destroyed them.
However, advances in AI may mean that this is no longer guaranteed, opening up the potential of taking out an enemy's ability to counter attack - and there may be no way to prevent such a future.
The claims were put forward by Rand Corporation, a non-profit based organisation in Santa Monica, California, that offers research and analysis to the United States armed forces on global policy issues.
It says that, in the coming decades, advances in AI could erode Mad's ability to keep nuclear nations in check.
Improved AI agents, combined with more sensor and open-source data, could convince countries that their nuclear capabilities are vulnerable.
That, the report says, may cause them to take more drastic steps to keep up with countires such as the US.
Speaking about it, Andrew Lohn, co-author on the paper and associate engineer at Rand said, “Some experts fear that an increased reliance on artificial intelligence can lead to new types of catastrophic mistakes.”
Another worrying scenario, the report claims, is that commanders could make decisions to launch strikes based on advice from AI assistants that have been fed the wrong information. The study is part of a broader effort to envision critical security challenges in the world of 2040, considering the effects of political, technological, social, and demographic trends that will shape those security challenges....