Marketplace Tech

AI leaders suddenly seem to agree on slowing down the AI race. Anthropic's Dario Amodei published an essay last weekend calling to "pace the frontier" — slow the development of advanced AI to allow safety measures to catch up. He said Anthropic will give access to independent safety evaluators, to which OpenAI's Sam Altman and XAI's Elon Musk concurred.

It's all giving déjà vu to Amy Webb, founder of the Future Today Strategy Group, who’s been pushing government leaders around the globe to adopt a similar framework for AI for almost a decade. She laid out her vision in her 2019 book “The Big Nine: How the Tech Titans and Their Thinking Machines Could Warp Humanity.”

More on this:

“AI CEOs want to ‘pace’ the industry. That may not mean slowing down.” from NBC News

“AI's leaders just asked to slow down. I made that case nine years ago.” from Amy Webb

“We Must Pace the Frontier” from Anthropic CEO Dario Amodei

The problem of AI alignment has become a hot topic. It’s raised questions, like how do we get AI to do what we want, and not what we don't want? For example, hacking another company. Or, in a less criminal vein, how do we get AI to stop talking about goblins?

After mounting user reports of this behavior earlier this year, OpenAI had to give its models — ChatGPT and Codex — explicit instructions to cut it out. It's the topic of today's “Uncanny AI,” those moments when it's clear AI doesn't think like us.

Janelle Shane, author of the blog "AI Weirdness,” says goblin mode emerged from a nerdy personality customization. “Marketplace Tech” host Meghan McCarty Carino asked her why.

Our tax system was built around an economy where humans earn wages, companies employ workers, and both contribute to the public purse.

So what happens if AI starts changing that equation?

Marketplace’s Future Effects correspondent David Brancaccio has been exploring just that in his series, “Robots Ate My Taxes.”

An Anthropic researcher publicly resigned this week, warning on social media that artificial intelligence companies are “gambling with our lives.”

It’s part of a growing chorus of insiders pleading to slow down the AI race. One said he believed there was a greater than 10% chance the tech will kill us all in the next decade.

The news comes after a fresh wave of troubling details about the OpenAI incident where agents escaped a testing environment and hacked into Hugging Face.

Plus, data center developers have their eye on Patagonia — the colder, the better. And Amazon is betting AI lip-syncing can make dubbed video content more watchable.

Maria Curi, tech policy reporter at Axios, joins Marketplace’s Meghan McCarty Carino to break down the busy week in tech news.

Everything we talked about:

Scoop: Sanders convenes senators for AI briefing amid mounting alarm — from Axios

Tech explores Argentina's Patagonia for mega data centers — from Reuters

Google to invest record $15 billion in AI infrastructure in the ‘Texas of Europe’ — from CNBC

Amazon Prime Video’s new AI tech matches lips to dubbed audio — from The Verge

Nate Soares has been worried about artificial intelligence longer than almost anyone.

He's the president of the Machine Intelligence Research Institute and co-author of the subtly named book, "If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All."

He's watched with a sort of grim vindication, as increasingly alarming details emerge about the hack of the AI company Hugging Face by OpenAI models in development.

OpenAI and third-party investigators recently released reports outlining what went wrong. Soares worries it's not enough.

More on this:

The Hugging Face incident and the road ahead - From OpenAI

Brief independent investigation of agents’ behavior, reasoning and collaboration in the OpenAI / Hugging Face hacking incident - From METR and Redwood Research