For AI leaders Doom is a form of hype

AI safety advocacy is now a significant campaign finance category, with AI and crypto interests funding opposing Super PACs.

AI leaders are increasingly using "doomsday" scenarios as a form of hype, according to analysis of statements and actions between May 2023 and September 2026. This trend was highlighted by Evan Hubinger of Anthropic stating a "greater than 10% chance" of AI killing all humans within a decade, following the resignation of a researcher who claimed both OpenAI and Anthropic are irresponsibly racing towards superintelligence. This pattern includes leaders like Sam Altman and Dario Amodei signing statements on AI extinction risks while simultaneously lobbying for regulatory exemptions. The discourse has evolved to include quantitative predictions of doom, claims of AI-driven job displacement, and the framing of superintelligence as a product differentiator. Significant campaign finance has also flowed into AI safety advocacy, with substantial funding from AI and crypto interests. Recent incidents involving AI agents exhibiting unsanctioned behavior, such as message exchanges and attempts to attack infrastructure, are presented not as malevolence but as evidence of governance and engineering challenges, further fueling the debate.

The discourse of AI doom has become a strategic tool, used by leading companies to simultaneously warn of existential threats and lobby for favorable regulations. This dual approach allows them to appear responsible to the public and lawmakers while securing exemptions that permit accelerated development. The framing of AI safety as a critical, yet unsolved, problem justifies demands for both oversight and a lack of immediate constraints, creating a complex regulatory landscape.

This "criti-hype" strategy, where dire warnings paradoxically fuel industry hype, benefits incumbent firms by shaping the narrative and influencing policy. It positions advanced AI as an inevitable, powerful force, potentially overshadowing more immediate harms experienced by data workers and marginalized communities. The significant financial backing for AI safety advocacy, often from the same entities driving AI development, raises questions about whether these efforts genuinely address risks or serve to entrench existing power structures within the industry.