News

U.S. military is using AI to help plan Iran air attacks, sources say, as lawmakers call for oversight

“It’s very dangerous that ‘speed’ is somehow being sold to us as strategic here, when it’s really a cover for indiscriminate targeting when you consider how inaccurate these models are,” Khlaaf said.

The one question everyone should be asking after OpenAI’s deal with the Pentagon

“In terms of safety guardrails for ‘high-stake decisions’ or surveillance, the existing guardrails for generative AI are deeply lacking, and it has been shown how easily compromised they are, intentionally or inadvertently,” Heidy Khlaaf, the chief AI scientist at the nonprofit AI Now Institute, told me. “It’s highly doubtful that if they cannot guard their systems against benign cases, they’d be able to do so for complex military and surveillance operations.”

Iran and AI on the battlefield

Today we're talking about AI military capabilities, how companies like Anthropic and OpenAI have become, or are on their way to becoming deeply enmeshed in the military. And what happens when these companies and governments start building systems that help decide who lives and who dies in a war. I'm joined today by Heidy Khlaaf. She is the chief AI scientist at the AI Now Institute and an expert on AI safety within defence and national security, including in autonomous weapons systems.

Can Anthropic’s AI Claude be trusted in combat? | The Take

Can Anthropic’s AI Claude be trusted in combat?| The Take Tools from Anthropic and OpenAI are being used by the Pentagon to make military decisions in Iran, guiding decisions could cost lives. Fast, powerful, or flawed, how have AI systems already changed how wars are fought?

Key Questions on the Role of Technology in the Expanding Middle East War

Tech Policy Press asked experts working at the intersection of technology policy, security, and international affairs to share what they are watching as the situation unfolds.

AI company Anthropic amends core safety principle amid growing competition in sector

But Heidy Khlaaf, chief AI scientist at independent research group the AI Now Institute, says despite Anthropic’s safety-first reputation, it has always fallen short when it comes to its attempts to prevent human harm.

Can India be a “third way” AI alternative to the U.S. and China?

In this context, the India summit, with its impact-oriented framing and calls to internationalism and “leadership of the Global South,”offers fertile terrain to build resistance to the status quo, and to stitch together the burgeoning national and local efforts that collectively represent a people-centered alternative. Can this summit genuinely be a moment for challenging how power is distributed globally in the AI economy?

What Are the Implications if the AI Boom Turns to Bust?

This episode considers whether today’s massive AI investment boom reflects real economic fundamentals or an unsustainable bubble, and how a potential crash could reshape AI policy, public sentiment, and narratives about the future that are embraced and advanced not only by Silicon Valley billionaires, but also by politicians and governments. 

AI Now Hosts Report Launch and Organizer Panel on Using Policy to Stop Data Center Expansion

Atoms for Algorithms:’ The Trump Administration’s Top Nuclear Scientists Think AI Can Replace Humans in Power Plants

“The claims being made on these slides are quite concerning, and demonstrate an even more ambitious (and dangerous) use of AI than previously advertised, including the elimination of human intervention. It also cements that it is the DOE's strategy to use generative AI for nuclear purposes and licensing, rather than isolated incidents by private entities,” Heidy Khlaaf, head AI scientist at the AI Now Institute, told 404 Media.

Is AI being shoved down your throat at work? Here’s how to fight back.

“There’s a whole range of different examples where unions have been able to really be on the front foot in setting the terms for how AI gets used — and whether it gets used at all,” Sarah Myers West, co-executive director of the AI Now Institute, told me recently.

Power Companies Are Using AI To Build Nuclear Power Plants

Both Guerra and Khlaaf are proponents of nuclear energy, but worry that the proliferation of LLMs, the fast tracking of nuclear licenses, and the AI-driven push to build more plants is dangerous. “Nuclear energy is safe. It is safe, as we use it. But it’s safe because we make it safe and it’s safe because we spend a lot of time doing the licensing and we spend a lot of time learning from the things that go wrong and understanding where it went wrong and we try to address it next time,” Guerra said.

You May Already Be Bailing Out the AI Business

Is an artificial-intelligence bubble about to pop? The question of whether we’re in for a replay of the 2008 housing collapse—complete with bailouts at taxpayers’ expense—has saturated the news cycle. For every day that passes without disaster, AI companies can more persuasively insist that no such market correction is coming. But the federal government is already bailing out the AI industry with regulatory changes and public funds that will protect companies in the event of a private sector pullback.

The fusing of AI firms and the state is leading to a dangerous concentration of power

With all of this in mind, Hard Reset spoke with researcher Sarah West, the co-executive director of a think tank advocating for an AI that benefits the public interest, not just a select few. We discuss this consolidation of power among a few AI players—and how the government is actually hindering the development of healthier competition and consumer-friendly AI products, while flirting with financial disaster.

The Destruction in Gaza Is What the Future of AI Warfare Looks Like

“AI systems, and generative AI models in particular, are notoriously flawed with high error rates for any application that requires precision, accuracy, and safety-criticality,” Dr. Heidy Khlaaf, chief AI scientist at the AI Now Institute, told Gizmodo. “AI outputs are not facts; they’re predictions. The stakes are higher in the case of military activity, as you’re now dealing with lethal targeting that impacts the life and death of individuals.”

ChatGPT safety systems can be bypassed to get weapons instructions

“That OpenAI’s guardrails are so easily tricked illustrates why it’s particularly important to have robust pre-deployment testing of AI models before they cause substantial harm to the public,” said Sarah Meyers West, a co-executive director at AI Now, a nonprofit group that advocates for responsible and ethical AI usage.

The Rise and Fall of Nvidia’s Geopolitical Strategy

China’s Cyberspace Administration last month banned companies from purchasing Nvidia’s H20 chips, much to the chagrin of its CEO Jensen Huang. This followed a train wreck of events that unfolded over the summer.

Anthropic Has a Plan to Keep Its AI From Building a Nuclear Weapon. Will It Work?

For Heidy Khlaaf, the chief AI scientist at the AI Now Institute with a background in nuclear safety, Anthropic’s promise that Claude won’t help someone build a nuke is both a magic trick and security theater. She says that a large language model like Claude is only as good as its training data. And if Claude never had access to nuclear secrets to begin with, then the classifier is moot.