Dario Strangelove, or: How I Learned to Stop Worrying and Love the AI Bomb

Recently, my wife and I have been rewatching classic television series. We noticed a striking theme in AMC's Mad Men, specifically in its season two finale. The finale featured Don Draper witnessing President Kennedy's announcement of Soviet missiles in Cuba. The characters contemplated their desired locations in the event of an attack.

While the crisis seemed irrelevant to the plot, it forced them to confront their deepest fears. This played a role in Don's recent disappearance and generated unusual behavior among the characters. The debate surrounding artificial intelligence safety shares similar concerns with those of the past generations. The constant threat of the end has left an indelible mark on society, much like the fears of the 1980s and 1960s.

The Berlin Wall's fall and the Cuban Missile Crisis were just a few examples of a society living on the edge of destruction for 75 years. The novel Dr. Strangelove or: How I Learned to Stop Worrying and Love the Bomb, a comedic portrayal of the atomic bomb, was released in 1964. Its release coincided with Tom Cruise's 2025 film Digger, which appears to consciously pay homage to Dr. Strangelove's over-the-top energy.

Tom Cruise's movie also seems to focus on the theme of artificial intelligence. CEO of Anthropic, Dario Amodei, is a known advocate for addressing the risks associated with AI development. He regularly hands out a copy of Leo Szilard's book, The Making of the Atomic Bomb, to new Anthropic employees. Amodei has expressed his concerns about the potential risks associated with AI models, such as loss of control, misuse for cyberattacks, and bioterrorism.

Amodei has proposed measures such as outside monitoring, coordination among labs, and international cooperation to address these risks. The AI safety debate is not a new phenomenon. Many of the concerns voiced today echo those of previous generations who lived through existential threats like nuclear annihilation. Janan Ganesh, a Financial Times columnist, has noted that the fears of past generations were often founded on large, concrete events, such as missile deployments, terrorist attacks, or pandemics. The constant 10% chance of AI-induced catastrophe has become a familiar refrain in Silicon Valley.