A.I. fear-mongering won't make the world a better place, but a less hopeful one
The rise of artificial intelligence is the industrial revolution of our times. Revolutions, by definition, upend existing norms, and novelty breeds fear. With every promise a transformative technology holds for improving society, comes the specter of the worst-case scenario. Will the robots wreak havoc and take over the world? Will they become so intelligent that they’re no longer under human control, or used to start wars and breed anarchy?
The short answer is: Certainly not with the technology we have today or what we can build in the foreseeable future. The longer one is more complex, but the nuance is critical for the long-term promise of A.I. to make lasting, positive impact across industries and our lives. There is undoubtedly a need for caution and vigilance when it comes to something this transformative. All revolutionary inventions need guardrails to ensure they help rather than hurt us–like the literal guardrails with high voltage warnings on train tracks. But contrary to those painting the bleakest picture of how A.I. might harm rather than help, there’s little evidence right now that the machine apocalypse is nigh. And its benefits will be increasingly obvious as this technology becomes universal in business and daily life.
We’re already marching toward that future. OpenAI’s ChatGPT chatbot can write up a personalized parody of a Seinfeld episode featuring your friends just as easily as it can track down and explain complex academic studies on medicine. Google, Microsoft, and a collection of much smaller tech companies (alongside independent coders using open-source algorithms) are harnessing A.I. to do everything from giving you a productivity boost at the office to generating new songs featuring artists who didn’t actually collaborate on the track. The rise of self-teaching machines isn’t just inevitable–it’s well underway.
The anxiety born of this radical shift isn’t surprising. Fear of progress isn’t new and in some cases may be warranted. But if history is a guide, the handwringing over A.I. will prove premature and misguided. Society has managed to adjust to transformative change on once unimaginable scales, like the invention of electricity. A.I. will be subject to that same scrutiny, its underlying technologies will be tweaked and improved, and any necessary regulations will follow–just the same as when cars reshaped the very foundation of personal mobility before we thought up safety measures like seatbelts and emergency brakes. (Relatedly: A.I. is once again transforming automobiles with smart, autonomous vehicles.) In the same way, we will eventually embrace A.I.’s positive impact on human life while fixing its flaws.
For society, artificial intelligence will prove the great equalizer of our times, a powerful tool to democratize services often siloed to the wealthy and fortunate few, while also uplifting human capacity to new heights. There’s no Spotify reserved for the upper class–it gives everyone access to music from around the world. “Computer” was originally a term for a person whose job was to perform calculations. They were replaced by spreadsheets on the machine computers of today, up-leveling humans to become analysts who provide context and actionable intelligence to numbers rather than just crunching them.
A.I. will pave the road for that same kind of mass access to myriad services critical to our lives (medicine, law, education), and augment our ability to perform the higher functions unique to humans. And it will only become smarter and more useful as the early iterations we see today continue to self-teach through machine learning.
We’ve already seen this kind of impact across multiple fields, including medicine. Why spend crucial physician time that could be used to treat more patients more efficiently on examining X-rays and MRIs when you could just use A.I. to identify a tumor or concerning abnormality? That’s not a rhetorical question. A.I. use in U.S. radiology departments shot up from zero in 2015 to 30% by 2020, facilitating faster initial diagnoses in diseases like cancer that can then be double-checked by a doctor.
Used responsibly with broad, unbiased information, A.I. could spur social change based on real-world evidence. In nations across the world, A.I. is already being used to assist in judicial investigations and boilerplate decision-making. Potentially, information collected by legal experts and watchdog groups could be fed into an A.I. system to identify concerning trends, for instance, in how minorities are treated under the law.
The more pessimistic fears are just new versions of problems that have existed in some form or another throughout human history. Scams evolved from the snake oil salesman at the door to mail to email. Human knowledge has always been inherently imperfect and a never-ending work in progress. Such relatively addressable concerns should not be used to bring down the regulatory hammer on A.I. in America. Hobbling A.I. at this critical moment won’t make the world a better place, but a less hopeful one.
You’d be hard-pressed to find many who think we’d be better off without electricity or cars. In 30 years, we’ll think the same about A.I.
Vijay Pande, Ph.D., is a general partner at Andreessen Horowitz, where he focuses on investments in biopharma and healthcare.
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.
