SyncAI.news, a Varaisys broadcasting
Huang and Zuckerberg back AI safety without a slowdown: Why Big Tech is betting on market forces to police AI
MA

Mint: AI

· 1 min read

IndiaMint: AI

Huang and Zuckerberg back AI safety without a slowdown: Why Big Tech is betting on market forces to police AI

The AI industry faces a divide over development pace, with leaders like Sam Altman advocating for a slowdown, while Jensen Huang and Mark Zuckerberg support continued progress, emphasising market incentives and internal safety measures for AI advancements.

The debate over at what pace AI should advance is creating a massive divide in the technology industry. While OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei and others are calling for a slowdown in the development of certain AI models, Nvidia CEO Jensen Huang and Meta CEO Mark Zuckerberg are backing a different approach. They say the industry should continue to push ahead, relying on market incentives and companies’ own safety measures to manage the risks.

Huang, during a recent conference, said that market forces already provide a mechanism that can keep AI companies in check, so new laws or regulations are not really required.

“It’s a false choice,” Huang told Salesforce CEO Marc Benioff, who was also present at the conference, arguing that companies could pursue both rapid development and safety at the same time.

However, he noted that AI companies should move quickly but pause when they believe a product is not safe.

“Run as fast as you can,” he said, adding that companies should stop and “make sure you get it right” if they feel a product or the company itself is out of control.

In a similar note, Meta boss Mark Zuckerberg said AI companies have strong commercial incentives to invest in safety and added that AI labs that fail to focus on “alignment”—work aimed at ensuring AI systems adhere to human values—could fall behind.

“There is a lot of debate about slowing progress on capabilities until alignment catches up,” Zuckerberg said. “My view is that trust and alignment are quickly becoming the most important capabilities that will differentiate agents and models.”

Because AI labs risk “significant liability if their models cause harm,” Zuckerberg believes

Sanchari Ghosh

Original source

This story was published by Mint: AI. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on livemint.com

Similar News