OpenAI’s CTO says government regulators should be ‘very involved’ in regulating AI

Mira Murati, the chief technology officer at OpenAI, believes government regulators should be “very involved” in developing safety standards for the deployment of advanced artificial intelligence models such as ChatGPT. 

She also believes a proposed six-month pause on development isn’t the right way to build safer systems and that the industry isn’t currently close to achieving artificial general intelligence (AGI) — a hypothetical intellectual threshold where an artificial agent is capable of performing any task requiring intelligence, including human-level cognition. Her comments stem from an interview with the Associated Press published on April 24.

Related: Elon Musk to launch truth-seeking artificial intelligence platform TruthGPT

When asked about the safety precautions OpenAI took before the launch of GPT-4, Murati explained that the company took a slow approach to training to not only inhibit the machine’s penchant for unwanted behavior but also to locate any downstream concerns associated with such changes:

“You have to be very careful because you might create some other imbalance. You have to constantly audit […] So then you have to adjust it again and be very careful about every time you make an intervention, seeing what else is being disrupted.”

In the wake of GPT-4’s launch, experts fearing the unknown-unknowns surrounding the future of AI have called for interventions ranging from increased government regulation to a six-month pause on global AI development.

The latter suggestion garnered attention and support from luminaries in the field of AI such as Elon Musk, Gary Marcus, and Eliezer Yudkowski, while many notable figures including Bill Gates, Yann LeCun and Andrew Ng have come out in opposition.

a big deal: @elonmusk, Y. Bengio, S. Russell, ⁦⁦@tegmark⁩, V. Kraknova, P. Maes, ⁦@Grady_Booch, ⁦@AndrewYang⁩, ⁦@tristanharris⁩ & over 1,000 others, including me, have called for a temporary pause on training systems exceeding GPT-4 https://t.co/PJ5YFu0xm9

— Gary Marcus (@GaryMarcus) March 29, 2023

For her part, Murati expressed support for the idea of increased government involvement, stating, “these systems should be regulated.” She continued: “At OpenAI, we’re constantly talking with governments and regulators and other organizations that are developing these systems to, at least at the company level, agree on some level of standards.”

But, on the subject of a developmental pause, Murati’s tone was…