Some days I buy into the AI as Normal Technology argument that Arvind Narayanan and Sayash Kapoor popularized in their essay by the same title.
Then I’ll read something like Anthropic’s S-1 filing, and the founder’s letter that precedes it. Anthropic is not pitching investors “AI as normal technology.” In the S-1 they describe ambitions for biology that go way beyond helping scientists write code and summarize papers.
On discoveries it says it has already made:
We have started to directly make biological discoveries with Claude: among them is the discovery of a molecular machine that might have the potential to be developed into a novel biotechnology tool, perhaps for gene editing.
On what it expects to follow:
First, we believe that AI will profoundly accelerate the rate of progress in biology and medicine, potentially curing diseases that have plagued humanity for millennia, including cancer, Alzheimer’s disease, heart disease, rare genetic diseases, and the major infectious diseases.
On biological misuse:
AI can be misused by those who wish to do harm, for example to conduct cyberattacks or to design biological weapons. We have strict guardrails and trusted access programs that are designed to block misuse of our technology while still enabling positive applications.
And on the incentives of the company developing it:
AI has unusually large externalities, both positive and negative. Cultivating the positive externalities, and defending against the negative ones, is to some degree in tension with the natural incentives of a for-profit company, and with the standard governance structure of a publicly traded company.
On economic disruption:
The incredible economic value of AI may come with economic disruption, similar to what has happened with previous technologies but greater in magnitude.
On empowering bad governments:
Authoritarian governments may use AI to create high-tech surveillance states, deepen repression, and make themselves much more difficult to dislodge.
It’s fine to be skeptical of some of their claims. But at the same time, take seriously what Anthropic is asking investors to believe: that the technology they’re developing could cure diseases and create biological weapons, could result in economic abundance or upheaval, could empower democratic and authoritarian governments alike. And that ordinary, normal technology commercial incentives are almost certainly inadequate to manage the consequences.
I don’t think this is the S-1 of a normal technology.
We believe that Al is one of the most consequential technologies in human history. If managed well, it could rapidly accelerate scientific innovation, greatly increase economic growth rates, bring cures for major diseases within reach, and solve problems that have resisted generations of human effort.