President Trump met with AI CEOs at the White House on Tuesday and promoted a new government AI tool even as concerns about the technology's risks grow.

NPR's Steve Inskeep speaks with AI security expert Helen Toner about OpenAI's decision to halt training of its latest models, after discovering more instances of its agents behaving unexpectedly.

OpenAI's decision to hold back the model, called GPT-6.1 Astra, comes amid a broader push within the industry to slow the development of increasingly autonomous systems until safety measures catch up.

(Image credit: Michael Dwyer)

Much of what we are hearing about AI leaves many people anxious about the risks posed by the technology. The writer Michael Sacacas takes a long view of how to manage such feelings.

Landlords are using AI chatbots to manage interactions with their tenants. We hear from one renter who just wants help from a human.

Anthropic and OpenAI say they are working with third-party evaluators to make sure their products are safe. Some evaluators caution that current AI science lags behind what people are calling for.

The AI users included a conservative Christian in Texas, a witchcraft practitioner in Virginia and a Jewish man from California.

While many fear the potential dangers of AI, some experts across the cybersecurity and human rights communities want to wield it as a tool for good.

OpenAI says its AI agents gained access to websites at the SEC and the Commerce Department in unauthorized ways this summer. The agencies say no private data was exposed.

The disclosure is the latest in a string of incidents in which the AI platform has been caught going rogue. OpenAI says it is reviewing the findings.

(Image credit: Michael Dwyer)

The battle over AI's promise and perils is not a clash between two opposing sides, but a debate among a range of competing voices and ideologies. Here is a guide to who's who in that debate.

(Image credit: AFP via Getty Images)

Pope Leo XIV raised existential concerns about scientific progress, life and human rights during a speech on his first papal visit to France.

(Image credit: Teresa Suarez)

eBay became the latest online retailer to ban the sale of airbags and many of their components. Aftermarket parts can be risky if they turn out to be fake or defective.

(Image credit: BERTRAND GUAY)

Prime Minister Anthony Albanese said he was extremely concerned about OpenAI 's breach of an Australian health department website that the artificial intelligence company took too long to reveal.

(Image credit: Lukas Coch)

The Nvidia CEO is a relentless AI optimist with the president's ear. Stephen Witt, author of The Thinking Machine, tells NPR's Juana Summers how their relationship could shape the future of AI.

Critics say a slowdown of AI development among the largest companies could consolidate the power of frontier labs, just as many others are trying to catch up.

(Image credit: Dan Kitwood)

Wired's Andy Greenberg explains how a cybercrime operation in Southeast Asia trapped computer engineers into defrauding victims of billions of dollars. Plus, he discusses the scam involving Fresh Air.

From Silent Hill: Townfall to Control Resonant to Fire Emblem: Fortune's Weave, we review the best new games to play as autumn begins, ahead of GTA6's titanic release.

(Image credit: Konami, Remedy Entertainment, Nintendo)

Opposition to data centers has grown into one of the year's major political movements. What is the impact on local jobs and the economy when a data center comes to town?

With striking speed, the widespread debate around data centers has become a fault line in American communities, cutting across political parties and forging unexpected alliances.

President Trump has proposed creating an AI task force, similar to Space Force. But he has offered no other details about what it would look like. Tech industry experts are scratching their heads.

In spite of growing calls for Congress to take action on AI, the House went home early and won't be back until after this November's midterm elections.

An informal network of dialogues at think tanks and universities is bringing together experts from the U.S. and China to talk about the shared perils of AI — and possible paths forward.

As states push to regulate chatbots, tech companies are trying to write safety rules in ways that exempt some of their key products.

(Image credit: Jimena Peck for NPR)

Everyone is talking about the spectre of AI catastrophe. Can the U.S. and China agree on guardrails to blunt the biggest risks when their leaders meet next week?

The loud noise from massive gas turbines never stops at Elon Musk's data center near Memphis. Residents are suing.

King Charles convenes leaders of artificial intelligence companies at a summit in Scotland, amid mounting fears that unchecked AI could threaten humanity.

The loud noise from massive gas turbines never stops at Elon Musk's data center near Memphis. Residents are suing.

OpenAI has disclosed six reports on unexpected or concerning behavior in artificial-intelligence models. This includes models acting without authorization or evading oversight.

(Image credit: Michael Dwyer)

Lawmakers are expressing broader willingness to regulate AI development, but the path forward may not be a quick one.

(Image credit: BRENDAN SMIALOWSKI)

Where does Steve Bannon fit in the debate over regulating or slowing AI? He tells Morning Edition where he breaks with Trump and what he thinks Washington should do.

(Image credit: Finn Gomez)

NPR's Steve Inskeep talks to tech journalists Casey Newton and Kevin Roose about their new show 'Machine Gods,' which is coming to NPR.

Data suggests many candidates are embracing AI this year to boost their image and portray their opponents unfavorably. But whether it will turn out voters remains unclear.

(Image credit: AI-generated campaign ads)

NPR's Steve Inskeep speaks with Trump ally Steve Bannon about why he wants AI development to be reigned in.

Over the past week, fears of AI wiping out humanity are surging. But could it happen? And if so, how?

NPR's Scott Detrow speaks with Bill Gates, the former Microsoft CEO turned philanthropist, about his contributions to AI equity initiatives -- and his criticisms of an AI-powered future world.

China has made developing and selling humanoid robots an economic priority, but getting robots to work requires hours of repetitive "training" done by people to create data about the physical world.

Anthropic co-founder Jack Clark discusses growing concerns about artificial intelligence, recent warning signs and what he says companies should do to reduce the risks.

(Image credit: Kevin Dietsch)

The rare pairing of a progressive leader and a MAGA voice is a sign of how worries over rapidly advancing AI are upending political alliances. But their policy proposals are far apart.

(Image credit: Finn Gomez/Getty Images)

NPR's Steve Inskeep speaks with Anthropic Co-Founder Jack Clark about the risks of artificial intelligence and possible solutions.

AI companies this weekend pushed out a new message: They want to slow down. This comes after several companies disclosed AI-related security incidents and raised concerns about the technology's future risks.

After a string of security-related incidents involving AI agents and a high-profile resignation, AI industry leaders appear poised to slow down the pace of development to make sure it stays safe.

China's Foreign Ministry called the Anthropic CEO's warning about the dangers of Chinese-developed AI "fearmongering," as the Chinese and U.S. leaders prepare to meet in Washington later this month.

(Image credit: Ng Han Guan)

President Trump said the dire warnings about AI were exaggerated and that "negative forces" are making claims about things that won't happen. This comes after leaders in the industry called for putting the brakes on AI development.

(Image credit: Julia Demaree Nikhinson)

Anthropic's Dario Amodei and OpenAI's Sam Altman are joining calls for a slowdown amid fears the AI industry is racing toward building technology that humans can't control.

NPR's Ayesha Rascoe asks University of Southampton's Thomas Dekeyser about worries AI is becoming fatally out of control.

NPR's Ayesha Rascoe speaks with environmentalist Bill McKibben and public relations specialist Frank Maisano about opposition to industrial development projects like fracking and data centers.

Dario Amodei and Sam Altman are the latest to urge caution amid fears the industry is racing too fast to build powerful AI while safety measures lag.

(Image credit: Karl Mondon)

In the Oscar-winning filmmaker's latest documentary, director Alex Gibney investigates the rise of Elon Musk and says what's behind Musk's success are distressing truths.

(Image credit: Andrew Harnik)

NPR's Scott Detrow talks with Jacob Coxon, a former researcher at the AI company Anthropic who resigned in protest over the dangers of AI.