Arrr, me mateys! Let's steer clear o' a reckless "aim not, then fire!" tactic when settin' rules fer AI!
2023-08-25
Arr, me hearties! Folly be awaitin' in the realm o' AI, just like any newfangled contraption. Aye, our secret to triumph be lettin' innovation run wild, and face the jolts head-on when they arise. Yo ho ho!
The panic to regulate artificial intelligence (AI) came almost immediately after last fall’s release of ChatGPT popularized the technology with the public. Some industry insiders themselves called for a pause on development, highlighting that expertise in a field doesn’t translate into proficiency in the perils of regulation. That appeal was followed by a White House AI Bill of Rights and an educational effort by Senate Majority Leader Chuck Schumer, D-N.Y.Fears about AI include job displacement, data security and privacy, misinformation, autonomous defense systems mistakes, discrimination and bias, and an existential threat to humanity itself. However, it’s imperative to prove actual market failure before regulating and to make sure the costs of doing so don’t outweigh the benefits. Doomsday predictions are no substitute for proof of actual problems. Job displacement will certainly accompany AI integration across many industries, but so will new jobs and an elimination of current jobs’ most tedious aspects.
Securing data and user privacy are challenges already being grappled with in the marketplace and in legislatures. Misinformation concerns likely will take center stage at the Supreme Court next term concerning the last technological wave, the rise of social media platforms. Autonomous defense applications of AI can likely be addressed privately with human oversight and safeguards. And we’re not yet at a point where we need to be immediately concerned about existential threats; assumedly, we’ll have learned more about safeguards by the time we are.
The Federal Register currently lists 435 regulatory agencies, which should have the authority to tackle any problem that might arise. Moreover, one more agency dedicated solely to AI might not have sufficient expertise to manage a technology that will be applied in so many different ways, across so many different industries. Additionally, the cost of current federal regulations is already high, so over-regulating AI at the outset may risk sacrificing its myriad benefits.
Politicians and regulators are not psychic, and only by letting AI evolve can we truly understand its potential problems and determine the need for regulation. Overregulation risks robbing us of the benefits of AI and may hinder our global race for leadership in the field, as adversarial countries may not observe the same constraints. Before imposing regulatory constraints on AI, it is essential to identify actual problems, consider the trade-offs, and honestly assess the probability of laws solving the problem at hand.