AI Products Should Not Be Treated as Independent Actors

Describing AI products as “superintelligence” or “fraud models” attributes agency to the product rather than to the company that manufactures it. This framework promotes these companies’ products as “superhuman” while simultaneously helping them avoid responsibility for their actions.

Instead of OpenAI being charged with creating malware that hacks other companies, press releases, news outlets, media personalities, and members of Congress refer to the “rogue model” as if it acted on its own. Instead of researchers questioning companies about their practices of plagiarizing academics’ work or using customer data to train models without consent, the public’s imagination is directed toward fears about what the future holds as hypothetical superintelligent machines emerge.

Data Centers Bring Real-World Costs

The AI industry is struggling with a popular bipartisan anti-data-center movement. It is a “distraction” from attempts to regulate the terrifying “superhuman” machines these companies are building. According to the AI industry, we have to worry more about a hypothetical god-in-the-machine than about the climate change these data centers are exacerbating, asthma suffered by people living nearby, increases in electricity costs and public subsidies, or water directed toward cooling.

Policymakers Should Approach AI Hype With Skepticism

We know it is better to make decisions based on evidence than to succumb to corporate pressure to make those decisions quickly. Wise decision-making by policymakers and communities requires time to hear from independent experts and put companies’ claims in context.

The best possible outcome from this summer’s hype is for policymakers and the public at large to take a breath, remain skeptical, and learn to recognize this kind of hype for what it is the next time it happens.