Subscriber Benefit
As a subscriber you can listen to articles at work, in the car, or while you work out. Subscribe Now
This week, Americans were confronted with a frightening question: What happens when artificial intelligence becomes powerful enough that humans struggle to control it?
AI researcher Jacob Coxon recently resigned from Anthropic, warning that leading AI companies are racing toward increasingly powerful systems without sufficient safeguards.
His concerns joined a growing conversation among researchers about the risks of increasingly autonomous AI. And now OpenAI has released GPT-6 Astra.
OpenAI President Greg Brockman has suggested that history might eventually look back at this period as the beginning of artificial general intelligence, or AGI.
Whether Astra meets that still-debated definition matters less to me than what these rapidly increasing capabilities mean for the rest of us.
Perhaps it is the timing of experiencing all of this while I am in law school, but I find myself reading these warnings differently. That should get our attention; it shouldn’t send us into a panic.
America cannot approach transformative technology from two equally dangerous extremes: pretending there is no risk or becoming so consumed by potential catastrophe that we regulate innovation out of existence.
There is a third option: Govern it.
Americans are right to ask hard questions about AI. But our apprehension does not have to become fearmongering. It can become a blueprint.
If we fear autonomous systems escaping human control, build human oversight into the law. If we fear cyberattacks, establish cybersecurity standards.
If we fear companies hiding failures, mandate disclosure. And if we fear no one will be responsible when AI causes harm, establish liability.
Our fears can tell us where the guardrails belong. We have done this before. Aviation, pharmaceuticals and nuclear technology brought extraordinary possibilities alongside extraordinary risks. America did not abandon innovation. We built laws, standards and institutions around it.
AI requires the same seriousness. But we do not need to regulate every use of AI as though it carries the same risk.
The AI helping your child with algebra is not the same as an autonomous system capable of discovering cybersecurity vulnerabilities. Regulate the capability and the risk.
The greater the capability, the greater the responsibility.
As AI systems become more powerful and autonomous, our regulatory expectations should rise with them. The most advanced systems should be subject to rigorous safety testing, independent audits, cybersecurity standards and mandatory incident reporting.
And as AI assumes greater responsibility for consequential decisions, the law must make equally clear who remains accountable when something goes wrong. AI cannot become a liability shield.
This is also where AI ethics must become more than a corporate mission statement. Ethics tells us what responsible innovation should look like. Compliance builds the systems to achieve it. Law establishes the floor below which no company may fall.
America does not have to choose between technological leadership and public safety. In fact, getting governance right might be essential to maintaining both.
Risk cannot always be eliminated. It can be governed. The emergence of increasingly autonomous artificial intelligence does not make law obsolete. It makes sophisticated technology governance indispensable.
The technology is evolving; now the law must evolve with it.•
__________
Yates is the former director of diversity for the Indiana Republican Party, a political commentator and a law degree candidate.
Click here for more Forefront columns.
Please enable JavaScript to view this content.
Editor's note: You can comment on IBJ stories by signing in to your IBJ account. If you have not registered, please sign up for a free account now. Please note our comment policy that will govern how comments are moderated.