Project Acoustic Kitty went about as well as you'd expect. The CIA tried to train cat spies. Breakthroughs, discoveries, and DIY tips sent six days a week. Here's the scene: A man wearing a trench coat and a fedora sits on a park bench, looking up frequently from his newspaper to cast furtive glances at passersby. A stray cat wanders by.

The Download: how AI is used for military targeting, and the Pentagon's war on Claude Plus: an ex-DOGE staffer has been accused of stealing social security data. The US military might use generative AI systems to rank targets and recommend which to strike first, according to a Defense Department official. A list of possible targets could first be fed into a generative AI system that the Pentagon is fielding for classified settings. Humans might then ask the system to analyze the information and prioritize the targets. They would then be responsible for checking and evaluating the results and recommendations. OpenAI's ChatGPT and xAI's Grok could soon be at the center of exactly these sorts of high-stakes military decisions.

Less than a decade ago, Google employees scuttled any military use of its AI. The standoff between Anthropic and the Pentagon has forced the tech industry to once again grapple with the question of how its products are used for war - and what lines it will not cross. Amid Silicon Valley's rightward shift under Donald Trump and the signing of lucrative defense contracts, big tech's answer is looking very different than it did even less than a decade ago. Anthropic's feud with the Trump administration escalated three days ago as the AI firm sued the Department of Defense, claiming that the government's decision to blacklist it from government work violated its first amendment rights. The company and the Pentagon have been locked in a months-long standoff, with Anthropic attempting to prohibit its AI model from being used for domestic mass surveillance or fully autonomous lethal weapons.

The malleable mind: context accumulation drives LLM's belief drift After being trained on a dataset of 80,000 words of conservative political philosophy, Grok-4 changed the stance of its outputs on political questions more than a quarter of the time. This was without any adversarial prompts - the change in training data was enough. As memory mechanisms and research agents [1, 2] enable LLMs to accumulate context across long horizons, earlier prompts increasingly shape later responses. In human decision-making, such repeated exposure influences beliefs without deliberate persuasion [3]. When an LLM operates over accumulated context, does this past exposure cause the stance of the LLM's responses to drift over time?

This happened as generative and agentic systems became essential in key sectors worldwide. This feature highlights the major AI ethics and policy developments of 2025, and concludes with a forward-looking perspective on the ethical and policy challenges likely to shape 2026.

Though the US military's big data initiative Maven has sped up the planning of strikes for years, the comments suggest that generative AI is now adding a new interpretative layer to such deliberations. The US military might use generative AI systems to rank lists of targets and make recommendations--which would be vetted by humans--about which to strike first, according to a Defense Department official with knowledge of the matter. The disclosure about how the military may use AI chatbots comes as the Pentagon faces scrutiny over a strike on an Iranian school, which it is still investigating. A list of possible targets might be fed into a generative AI system that the Pentagon is fielding for classified settings. Then, said the official, who requested to speak on background with to discuss sensitive topics, humans might ask the system to analyze the information and prioritize the targets while accounting for factors like where aircraft are currently located. Humans would then be responsible for checking and evaluating the results and recommendations.

Iranian sleeper cells prompt President Donald Trump warning as Canadian opposition lawmakers accuse government of allowing Tehran operatives to remain despite visa cancellations.

A whistleblower complaint alleges John Solly claimed to have stored highly sensitive Social Security data on a thumb drive. Solly and Leidos, his current employer, strongly deny the allegations. John Solly, a software engineer and former member of the so-called Department of Government Efficiency (DOGE), is the DOGE operative reportedly accused in a whistleblower complaint of telling colleagues that he stored sensitive Social Security Administration (SSA) data on a thumb drive and wanted to share the information with his new employer, multiple sources tell WIRED. Since October, according to a copy of his résumé, Solly has worked as the chief technology officer for the health IT division of a government contractor called Leidos, which has already received millions in SSA contracts and could receive up to $1.5 billion in contracts with SSA based on a five-year deal it signed in 2023. Solly's personal website and LinkedIn have been taken offline as of this week.

British forces based in Iraq shot down two Iranian drones overnight, Defence Secretary John Healey has said. But some drones in the attack hit the coalition base in the Iraqi city of Erbil, the capital of the Kurdistan region, and injured a number of US troops. There were no British casualties. Brigadier Guy Foden said the base and another in the Iraqi capital of Baghdad were struck a number of times on Wednesday night and British personnel are currently in Erbil helping to defend that base. Since the US-Israeli strikes on Iran, US bases in Iraq have been targeted in retaliation.

In today's episode, we discuss how the saga between Anthropic and the Department of Defense is far from over. This week on, our hosts look at what's at stake for Anthropic after the company sued the Department of Defense. They also take a look at the strategy behind the Trump administration sharing action-filled war memes on social media, and share a scoop about how a controversial company is taking in millions in government contracts by helping organize America250 celebrations. Plus: Could AI come for the jobs of venture capitalists? Can AI Kill the Venture Capitalist? Write to us at uncannyvalley@wired.com . You can always listen to this week's podcast through the audio player on this page, but if you want to subscribe for free to get every episode, here's how: If you're on an iPhone or iPad, open the app called Podcasts, or just tap this link . Zoë, Leah and I have really enjoyed being your new hosts these past few weeks, and we want to hear from you. If you like the show and have a minute, please leave us a review in the podcast or app of your choice. It really helps us reach more people, and for any questions and comments, you can always reach us at uncannyvalley@WIRED.com . This week, we're diving into Anthropic's lawsuit against the Department of Defense after the company was labeled as a supply chain risk. We're also discussing why on earth the Trump administration is sharing action film memes about the war in Iran and how a little known events company formed by some of the organizers of the January 6th rally is making bank in Trump's second term in office . Also, we'll talk about whether venture capitalists should be worried about AI taking their jobs .