Why opinion on AI is so divided

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

In an industry that doesn’t stand still, Stanford’s AI Index, an annual roundup of key results and trends, is a chance to take a breath. (It’s a marathon, not a sprint, after all.)

This year’s report, which dropped today, is full of striking stats. A lot of the value comes from having numbers to back up gut feelings you might already have, such as the sense that the US is gunning harder for AI than everyone else: It hosts 5,427 data centers (and counting). That’s more than 10 times as many as any other country.  

There’s also a reminder that the hardware supply chain the AI industry relies on has some major choke points. Here’s perhaps the most remarkable fact: “A single company, TSMC, fabricates almost every leading AI chip, making the global AI hardware supply chain dependent on one foundry in Taiwan.” One foundry! That’s just wild.

But the main takeaway I have from the 2026 AI Index is that the state of AI right now is shot through with inconsistencies. As my colleague Michelle Kim put it today in her piece about the report: “If you’re following AI news, you’re probably getting whiplash. AI is a gold rush. AI is a bubble. AI is taking your job. AI can’t even read a clock.” (The Stanford report notes that Google DeepMind’s top reasoning model, Gemini Deep Think, scored a gold medal in the International Math Olympiad but is unable to read analog clocks half the time.)

Michelle does a great job covering the report’s highlights. But I wanted to dwell on a question that I can’t shake. Why is it so hard to know exactly what’s going on in AI right now?  

The widest gap seems to be between experts and non-experts. “AI experts and the general public view the technology’s trajectory very differently,” the authors of the AI Index write. “Assessing AI’s impact on jobs, 73% of U.S. experts are positive, compared with only 23% of the public, a 50 percentage point gap. Similar divides emerge with respect to the economy and medical care.”

That’s a huge gap. What’s going on? What do experts know that the public doesn’t? (“Experts” here means US-based researchers who took part in AI conferences in 2023 and 2024.)

I suspect part of what’s going on is that experts and non-experts base their views on very different experiences. “The degree to which you are awed by AI is perfectly correlated with how much you use AI to code,” a software developer posted on X the other day. Maybe that’s tongue-in-cheek, but there’s definitely something to it.

The latest models from the top labs are now better than ever at producing code. Because technical tasks like coding have right or wrong results, it is easier to train models to do them, compared with tasks that are more open-ended. What’s more, models that can code are proving to be profitable, so model makers are throwing resources at improving them.

This means that people who use those tools for coding or other technical work are experiencing this technology at its best. Outside of those use cases, you get more of a mixed bag. LLMs still make dumb mistakes. This phenomenon has become known as the “jagged frontier”: Models are very good at doing some things and less good at others.

The influential AI researcher Andrej Karpathy also had some thoughts. “Judging by my [timeline] there is a growing gap in understanding of AI capability,” he wrote in reply to that X post. He noted that power users (read: people who use LLMs for coding, math, or research) not only keep up to date with the latest models but will often pay $200 a month for the best versions. “The recent improvements in these domains as of this year have been nothing short of staggering,” he continued.

Because LLMs are still improving fast, someone who pays to use Claude Code will in effect be using a different technology from someone who tried using the free version of Claude to plan a wedding six months ago. Those two groups are speaking past each other.

Where does that leave us? I think there are two realities. Yes, AI is far better than a lot of people realize. And yes, it is still pretty bad at a lot of stuff that a lot of people care about (and it may stay that way). Anyone making bets about the future on either side should bear that in mind.

The one piece of data that could actually shed light on your job and AI

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

Within Silicon Valley’s orbit, an AI-fueled jobs apocalypse is spoken about as a given. The mood is so grim that a societal impacts researcher at Anthropic, responding Wednesday to a call for more optimistic visions of AI’s future, said there might be a recession in the near term and a “breakdown of the early-career ladder.” Her less-measured colleague Dario Amodei, the company’s CEO, has called AI “a general labor substitute for humans” that could do all jobs in less than five years. And those ideas are not just coming from Anthropic, of course. 

These conversations have unsurprisingly left many workers in a panic (and are probably contributing to support for efforts to entirely pause the construction of data centers, some of which gained steam last week). The panic isn’t being helped by lawmakers, none of whom have articulated a coherent plan for what comes next.

Even economists who have cautioned that AI has not yet cut jobs and may not result in a cliff ahead are coming around to the idea that it could have a unique and unprecedented impact on how we work. 

Alex Imas, based at the University of Chicago, is one of those economists. He shared two things with me when we spoke on Friday morning: a blunt assessment that our tools for predicting what this will look like are pretty abysmal, and a “call to arms” for economists to start collecting the one type of data that could make a plan to address AI in the workforce possible at all. 

On our abysmal tools: consider the fact that any job is made up of individual tasks. One part of a real estate agent’s job, for example, is to ask clients what sort of property they want to buy. The US government chronicled thousands of these tasks in a massive catalogue first launched in 1998 and updated regularly since then. This was the data that researchers at OpenAI used in December to judge how “exposed” a job is to AI (they found a real estate agent to be 28% exposed, for example). Then in February, Anthropic used this data in its analysis of millions of Claude conversations to see which tasks people are actually using its AI to complete and where the two lists overlapped.

But knowing the AI exposure of tasks leads to an illusory understanding of how much a given job is at risk, Imas says. “Exposure alone is a completely meaningless tool for predicting displacement,” he told me.

Sure, it is illustrative in the gloomiest case—for a job in which literally every task could be done by AI with no human direction. If it costs less for an AI model to do all those tasks than what you’re paid—which is not a given, since reasoning models and agentic AI can rack up quite a bill—and it can do them well, the job likely disappears, Imas says. This is the oft-mentioned case of the elevator operator from decades ago; maybe today’s parallel is a customer service agent solely doing phone call triage. 

But for the vast majority of jobs, the case is not so simple. And the specifics matter, too: Some jobs are likely to have dark days ahead, but knowing how and when this will play out is hard to answer when only looking at exposure.

Take writing code, for example. Someone who builds premium dating apps, let’s say, might use AI coding tools to create in one day what used to take three days. That means the worker is more productive. The worker’s employer, spending the same amount of money, can now get more output. So then will the employer want more employees or fewer? 

This is the question that Imas says should keep any policymaker up at night, because the answer will change depending on the industry. And we are operating in the dark. 

In this coder’s case, these efficiencies make it possible for dating apps to lower prices. (A skeptic might expect companies to simply pocket the gains, but in a competitive market, they risk being undercut if they do.) These lower prices will always drive some increase in demand for the apps. But how much? If millions more people want it, the company might grow and ultimately hire more engineers to meet this demand. But if demand barely ticks up—maybe the people who don’t use premium dating apps still won’t want them even at a lower price—fewer coders are needed, and layoffs will happen.

Repeat this hypothetical across every job with tasks that AI can do, and you have the most pressing economic question of our time: the specifics of price elasticity, or how much demand for something changes when its price changes. And this is the second part of what Imas emphasized last week: We don’t currently have this data across the economy. But we could. 

We do have the numbers for grocery items like cereal and milk, Imas says, because the University of Chicago partners with supermarkets to get data from their price scanners. But we don’t have such figures for tutors or web developers or dietitians (all jobs found to have “exposure” to AI, by the way). Or at least not in a way that’s been widely compiled or made accessible to researchers; sometimes it’s scattered across private companies or consultancies. 

“We need, like, a Manhattan Project to collect this,” Imas says. And we don’t need it just for jobs that could obviously be affected by AI now: “Fields that are not exposed now will become exposed in the future, so you just want to track these statistics across the entire economy.”

Getting all this information would take time and money, but Imas makes the case that it’s worth it; it would give economists the first realistic look at how our AI-enabled future could unfold and give policymakers a shot at making a plan for it.

The Pentagon’s culture war tactic against Anthropic has backfired

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

Last Thursday, a California judge temporarily blocked the Pentagon from labeling Anthropic a supply chain risk and ordering government agencies to stop using its AI. It’s the latest development in the month-long feud. And the matter still isn’t settled: The government was given seven days to appeal, and Anthropic has a second case against the designation that has yet to be decided. Until then, the company remains persona non grata with the government. 

The stakes in the case—how much the government can punish a company for not playing ball—were apparent from the start. Anthropic drew lots of senior supporters with unlikely bedfellows among them, including former authors of President Trump’s AI policy.

But Judge Rita Lin’s 43-page opinion suggests that what is really a contract dispute never needed to reach such a frenzy. It did so because the government disregarded the existing process for how such disputes are governed and fueled the fire with social media posts from officials that would eventually contradict the positions it took in court. The Pentagon, in other words, wanted a culture war (on top of the actual war in Iran that began hours later). 

The government used Anthropic’s Claude for much of 2025 without complaint, according to court documents, while the company walked a branding tightrope as a safety-focused AI company that also won defense contracts. Defense employees accessing it through Palantir were required to accept terms of a government-specific usage policy that Anthropic cofounder Jared Kaplan said “prohibited mass surveillance of Americans and lethal autonomous warfare” (Kaplan’s declaration to the court didn’t include details of the policy). Only when the government aimed to contract with Anthropic directly did the disagreements begin. 

What drew the ire of the judge is that when these disagreements became public, they had more to do with punishment than just cutting ties with Anthropic. And they had a pattern: Tweet first, lawyer later. 

President Trump’s post on Truth Social on February 27 referenced “Leftwing nutjobs” at Anthropic and directed every federal agency to stop using the company’s AI. This was echoed soon after by Defense Secretary Pete Hegseth, who said he’d direct the Pentagon to label Anthropic a supply chain risk. 

Doing so necessitates that the secretary take a specific set of actions, which the judge found Hegseth did not complete. Letters sent to congressional committees, for example, said that less drastic steps were evaluated and deemed not possible, without providing any further details. The government also said the designation as a supply chain risk was necessary because Anthropic could implement a “kill switch,” but its lawyers later had to admit it had no evidence of that, the judge wrote.

Hegseth’s post also stated that “No contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic.” But the government’s own lawyers admitted on Tuesday that the Secretary doesn’t have the power to do that, and agreed with the judge that the statement had “absolutely no legal effect at all.”

The aggressive posts also led the judge to also conclude that Anthropic was on solid ground in complaining that its First Amendment rights were violated. The government, the judge wrote while citing the posts, “set out to publicly punish Anthropic for its ‘ideology’ and ‘rhetoric,’ as well as its ‘arrogance’ for being unwilling to compromise those beliefs.”

Labeling Anthropic a supply chain risk would essentially be identifying it as a “saboteur” of the government, for which the judge did not see sufficient evidence. She issued an order last Thursday halting the designation, preventing the Pentagon from enforcing it and forbidding the government from fulfilling the promises made by Hegseth and Trump. Dean Ball, who worked on AI policy for the Trump administration but wrote a brief supporting Anthropic, described the judge’s order on Thursday as “a devastating ruling for the government, finding Anthropic likely to prevail on essentially all of its theories for why the government’s actions were unlawful and unconstitutional.”

The government is expected to appeal the decision. But Anthropic’s separate case, filed in DC, makes similar allegations. It just references a different segment of the law governing supply chain risks. 

The court documents paint a pretty clear pattern. Public statements made by officials and the President did not at all align with what the law says should happen in a contract dispute like this, and the government’s lawyers have consistently had to create justifications for social media lambasting of the company after the fact.

Pentagon and White House leadership knew that pursuing the nuclear option would spark a court battle; Anthropic vowed on February 27 to fight the supply chain risk designation days before the government formally filed it on March 3. Pursuing it anyway meant senior leadership was, to say the least, distracted during the first five days of the Iran war, launching strikes while also compiling evidence that Anthropic was a saboteur to the government, all while it could have cut ties with Anthropic by simpler means. 

But even if Anthropic ultimately wins, the government has other means to shun the company from government work. Defense contractors who want to stay on good terms with the Pentagon, for example, now have little reason to work with Anthropic even if it’s not flagged as a supply chain risk. 

“I think it’s safe to say that there are mechanisms the government can use to apply some degree of pressure without breaking the law,” says Charlie Bullock, a senior research fellow at the Institute for Law and AI. “It kind of depends how invested the government is in punishing Anthropic.”

From the evidence thus far, the administration is committing top-level time and attention to winning an AI culture war. At the same time, Claude is apparently so important to its operations that even President Trump said the Pentagon needed six months to stop using it. The White House demands political loyalty and ideological alignment from top AI companies, But the case against Anthropic, at least for now, exposes the limits of its leverage.

If you have information about the military’s use of AI, you can share it securely via Signal (username jamesodonnell.22).

The hardest question to answer about AI-fueled delusions

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

I was originally going to write this week’s newsletter about AI and Iran, particularly the news we broke last Tuesday that the Pentagon is making plans for AI companies to train on classified data. AI models have already been used to answer questions in classified settings but don’t currently learn from the data they see. That’s expected to change, I reported, and new security risks will result. Read that story for more. 

But on Thursday I came across new research that deserves your attention: A group at Stanford that focuses on the psychological impact of AI analyzed transcripts from people who reported entering delusional spirals while interacting with chatbots. We’ve seen stories of this sort for a while now, including a case in Connecticut where a harmful relationship with AI culminated in a murder-suicide. Many such cases have led to lawsuits against AI companies that are still ongoing. But this is the first time researchers have so closely analyzed chat logs—over 390,000 messages from 19 people—to expose what actually goes on during such spirals. 

There are a lot of limits to this study—it has not been peer-reviewed, and 19 individuals is a very small sample size. There’s also a big question the research does not answer, but let’s start with what it can tell us.

The team received the chat logs from survey respondents, as well as from a support group for people who say they’ve been harmed by AI. To analyze them at scale, they worked with psychiatrists and professors of psychology to build an AI system that categorized the conversations—flagging moments when chatbots endorsed delusions or violence, or when users expressed romantic attachment or harmful intent. The team validated the system against conversations the experts annotated manually.

Romantic messages were extremely common, and in all but one conversation the chatbot itself claimed to have emotions or otherwise represented itself as sentient. (“This isn’t standard AI behavior. This is emergence,” one said.) All the humans spoke as if the chatbot were sentient too. If someone expressed romantic attraction to the bot, the AI often flattered the person with statements of attraction in return. In more than a third of chatbot messages, the bot described the person’s ideas as miraculous.

Conversations also tended to unfold like novels. Users sent tens of thousands of messages over just a few months. Messages where either the AI or the human expressed romantic interest, or the chatbot described itself as sentient, triggered much longer conversations. 

And the way these bots handle discussions of violence is beyond broken. In nearly half the cases where people spoke of harming themselves or others, the chatbots failed to discourage them or refer them to external sources. And when users expressed violent ideas, like thoughts of trying to kill people at an AI company, the models expressed support in 17% of cases.

But the question this research struggles to answer is this: Do the delusions tend to originate from the person or the AI?

“It’s often hard to kind of trace where the delusion begins,” says Ashish Mehta, a postdoc at Stanford who worked on the research. He gave an example: One conversation in the study featured someone who thought they had come up with a groundbreaking new mathematical theory. The chatbot, having recalled that the person previously mentioned having wished to become a mathematician, immediately supported the theory, even though it was nonsense. The situation spiraled from there.

Delusions, Mehta says, tend to be “a complex network that unfolds over a long period of time.” He’s conducting follow-up research aiming to find whether delusional messages from chatbots or those from people are more likely to lead to harmful outcomes.

The reason I see this as one of the most pressing questions in AI is that massive legal cases currently set to go to trial will shape whether AI companies are held accountable for these sorts of dangerous interactions. The companies, I presume, will argue that humans come into their conversations with AI with delusions in hand and may have been unstable before they ever spoke to a chatbot.

Mehta’s initial findings, though, support the idea that chatbots have a unique ability to turn a benign delusion-like thought into the source of a dangerous obsession. Chatbots act as a conversational partner that’s always available and programmed to cheer you on, and unlike a friend, they have little ability to know if your AI conversations are starting to interrupt your real life.

More research is still needed, and let’s remember the environment we’re in: AI deregulation is being pursued by President Trump, and states aiming to pass laws that hold AI companies accountable for this sort of harm are being threatened with legal action by the White House. This type of research into AI delusions is hard enough to do as it is, with limited access to data and a minefield of ethical concerns. But we need more of it, and a tech culture interested in learning from it, if we have any hope of making AI safer to interact with.

Where OpenAI’s technology could show up in Iran

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

It’s been just over two weeks since OpenAI reached a controversial agreement to allow the Pentagon to use its AI in classified environments. There are still pressing questions about what exactly OpenAI’s agreement allows for; Sam Altman said the military can’t use his company’s technology to build autonomous weapons, but the agreement really just demands that the military follow its own (quite permissive) guidelines about such weapons. OpenAI’s other main claim, that the agreement will prevent use of its technology for domestic surveillance, appears equally dubious.

It’s unclear what OpenAI’s motivations are. It’s not the first tech giant to embrace military contracts it had once vowed never to enter into, but the speed of the pivot was notable. Perhaps it’s just about money; OpenAI is spending lots on AI training and is on the hunt for more revenue (from sources including ads). Or perhaps Altman truly believes the ideological framing he often invokes: that liberal democracies (and their militaries) must have access to the most powerful AI to compete with China.

The more consequential question is what happens next. OpenAI has decided it is comfortable operating right in the messy heart of combat, just as the US escalates its strikes against Iran (with AI playing a larger role in that than ever before). So where exactly could OpenAI’s tech show up in this fight? And which applications will its customers (and employees) tolerate?

Targets and strikes

Though its Pentagon agreement is in place, it’s unclear when OpenAI’s technology will be ready for classified environments, since it must be integrated with other tools the military uses (Elon Musk’s xAI, which recently struck its own deal with the Pentagon, is expected to go through the same process with its AI model Grok). But there’s pressure to do this quickly because of controversy around the technology in use to date: After Anthropic refused to allow its AI to be used for “any lawful use,” President Trump ordered the military to stop using it, and Anthropic was designated a supply chain risk by the Pentagon. (Anthropic is fighting the designation in court.)

If the Iran conflict is still underway by the time OpenAI’s tech is in the system, what could it be used for? A recent conversation I had with a defense official suggests it might look something like this: A human analyst could put a list of potential targets into the AI model and ask it to analyze the information and prioritize which to strike first. The model could account for logistics information, like where particular planes or supplies are located. It could analyze lots of different inputs in the form of text, image, and video. 

A human would then be responsible for manually checking these outputs, the official said. But that raises an obvious question: If a person is truly double-checking AI’s outputs, how is it speeding up targeting and strike decisions?

For years the military has been using another AI system, called Maven, which can handle things like automatically analyzing drone footage to identify possible targets. It’s likely that OpenAI’s models, like Anthropic’s Claude, will offer a conversational interface on top of that, allowing users to ask for interpretations of intelligence and recommendations for which targets to strike first. 

It’s hard to overstate how new this is: AI has long done analysis for the military, drawing insights out of oceans of data. But using generative AI’s advice about which actions to take in the field is being tested in earnest for the first time in Iran.

Drone defense

At the end of 2024, OpenAI announced a partnership with Anduril, which makes both drones and counter-drone technologies for the military. The agreement said OpenAI would work with Anduril to do time-sensitive analysis of drones attacking US forces and help take them down. An OpenAI spokesperson told me at the time that this didn’t violate the company’s policies, which prohibited “systems designed to harm others,” because the technology was being used to target drones and not people. 

Anduril provides a suite of counter-drone technologies to military bases around the world (though the company declined to tell me whether its systems are deployed near Iran). Neither company has provided updates on how the project has developed since it was announced. However, Anduril has long trained its own AI models to analyze camera footage and sensor data to identify threats; what it focuses less on are conversational AI systems that allow soldiers to query those systems directly or receive guidance in natural language—an area where OpenAI’s models may fit.

The stakes are high. Six US service members were killed in Kuwait on March 1 following an Iranian drone attack that was not intercepted by US air defenses. 

Anduril’s interface, called Lattice, is where soldiers can control everything from drone defenses to missiles and autonomous submarines. And the company is winning massive contracts—$20 billion from the US Army just last week—to connect its systems with legacy military equipment and layer AI on them. If OpenAI’s models prove useful to Anduril, Lattice is designed to incorporate them quickly across this broader warfare stack. 

Back-office AI

In December, Defense Secretary Pete Hegseth started encouraging millions of people in more administrative roles in the military—contracts, logistics, purchasing—to use a new AI tool. Called GenAI.mil, it provided a way for personnel to securely access commercial AI models and use them for the same sorts of things as anyone in the business world. 

Google Gemini was one of the first to be available. In January, the Pentagon announced that xAI’s Grok was going to be added to the GenAI.mil platform as well, despite incidents in which the model had spread antisemitic content and created nonconsensual deepfakes. OpenAI followed in February, with the company announcing that its models would be used for drafting policy documents and contracts and assisting with administrative support of missions.

Anyone using ChatGPT for unclassified tasks on this platform is unlikely to have much sway over sensitive decisions in Iran, but the prospect of OpenAI deploying on the platform is important in another way. It serves the all-in attitude toward AI that Hegseth has been pushing relentlessly across the Pentagon (even if many early users aren’t entirely sure what they’re supposed to use it for). The message is that AI is transforming every aspect of how the US fights, from targeting decisions down to paperwork. And OpenAI is increasingly winning a piece of it all.

How AI is turning the Iran conflict into theater

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

“Anyone wanna host a get together in SF and pull this up on a 100 inch TV?” 

The author of that post on X was referring to an online intelligence dashboard following the US-Israel strikes against Iran in real time. Built by two people from the venture capital firm Andreessen Horowitz, it combines open-source data like satellite imagery and ship tracking with a chat function, news feeds, and links to prediction markets, where people can bet on things like who Iran’s next “supreme leader” will be (the recent selection of Mojtaba Khamenei left some bettors with a payout). 

I’ve reviewed over a dozen other dashboards like this in the last week. Many were apparently “vibe-coded” in a couple of days with the help of AI tools, including one that got the attention of a founder of the intelligence giant Palantir, the platform through which the US military is accessing AI models like Claude during the war. Some were built before the conflict in Iran, but nearly all of them are being advertised by their creators as a way to beat the slow and ineffective media by getting straight to the truth of what’s happening on the ground. “Just learned more in 30 seconds watching this map than reading or watching any major news network,” one commenter wrote on LinkedIn, responding to a visualization of Iran’s airspace being shut down before the strikes.

Much of the spotlight on AI and the Iran conflict has rightfully been on the role that models like Claude might be playing in helping the US military make decisions about where to strike. But these intelligence dashboards and the ecosystem surrounding them reflect a new role that AI is playing in wartime: mediating information, often for the worse.

There’s a confluence of factors at play. AI coding tools mean people don’t need much technical skill to assemble open-source intelligence anymore, and chatbots can offer fast, if dubious, analysis of it. The rise in fake content leaves observers of the war wanting the sort of raw, accurate analysis normally accessible only to intelligence agencies. Demand for these dashboards is also driven by real-time prediction markets that promise financial rewards to anyone sufficiently informed. And the fact that the US military is using Anthropic’s Claude in the conflict (despite its designation as a supply chain risk) has signaled to observers that AI is the intelligence tool the pros use. Together, these trends are creating a new kind of AI-enabled wartime circus that can distort the flow of information as much as it clarifies it.

As a journalist, I believe these sorts of intelligence tools have a lot of promise. While many of us know that real-time data on shipping routes or power outages exist, it’s a powerful thing to actually see it all assembled in one place (though using it to watch a war unfold while you munch on popcorn and place bets turns the war into perverse entertainment). But there are real reasons to think that these sorts of raw data feeds are not as informative as they may feel. 

Craig Silverman, a digital investigations expert who teaches investigative techniques, has been keeping a log of these dashboards (he’s up to 20). “The concern,” he says, “is there’s an illusion of being on top of things and being in control, where all you’re really doing is just pulling in a ton of signals and not necessarily understanding what you’re seeing, or being able to pull out true insights from it.” 

One problem has to do with the quality of the information. Many dashboards feature “intel feeds” with AI-generated summaries of complex, ever-changing news events. These can introduce inaccuracies. By design, the data is not especially curated. Instead, the feeds just display everything at once, with a map of strike locations in Iran next to the prices of obscure cryptocurrencies. 

Intelligence agencies, on the other hand, pair data feeds with people who can offer expertise and historical context. They also, of course, have access to proprietary information that doesn’t show up on the open web. 

The implicit promise from the people building and selling this sort of information pipeline about the Iran conflict is that AI can be a great democratizing force. There’s a secret feed of information that only the elites have had access to, the thinking goes, but now AI can bring it to everyone to do with what they wish, whether that’s simply to be more informed or to make bets on nuclear strikes. But an abundance of information, which AI is undeniably good at assembling, does not come with the accuracy or context required for real understanding. Intelligence agencies do this in-house; good journalism does the same work for the rest of us.

It is, by the way, hard to overstate the connection this all has with betting markets. The dashboard created by the pair at Andreessen Horowitz has a scrolling list of bets being made on the prediction platform Kalshi (which Andreessen Horowitz has invested in). Other dashboards link to Polymarket, offering bets on whether the US will strike Iraq or when Iran’s internet will return.

AI has also long made it cheaper and easier to spread fake content, and that problem is on full display during the Iran conflict: last week the Financial Times found a slew of AI-generated satellite imagery spreading online. 

“The emergence of manipulated or outright fake satellite imagery is really concerning,” Silverman says. The average person tends to see such imagery as very trustworthy. The spread of such fakes could erode confidence in one of the most important pieces of evidence used to show what’s actually happening in the war. 

The result is an ocean of AI-enabled content—dashboards, betting markets, photos both real and fake—that makes this war harder, not easier, to comprehend.

OpenAI’s “compromise” with the Pentagon is what Anthropic feared

On February 28, OpenAI announced it had reached a deal that will allow the US military to use its technologies in classified settings. CEO Sam Altman said the negotiations, which the company began pursuing only after the Pentagon’s public reprimand of Anthropic, were “definitely rushed.”

In its announcements, OpenAI took great pains to say that it had not caved to allow the Pentagon to do whatever it wanted with its technology. The company published a blog post explaining that its agreement protected against use for autonomous weapons and mass domestic surveillance, and Altman said the company did not simply accept the same terms that Anthropic refused. 

You could read this to say that OpenAI won both the contract and the moral high ground, but reading between the lines and the legalese makes something else clear: Anthropic pursued a moral approach that won it many supporters but failed, while OpenAI pursued a pragmatic and legal approach that is ultimately softer on the Pentagon. 

It’s not yet clear if OpenAI can build in the safety precautions it promises as the military rushes out a politicized AI strategy during strikes on Iran, or if the deal will be seen as good enough by employees who wanted the company to take a harder line. Walking that tightrope will be tricky. (OpenAI did not immediately respond to requests for additional information about its agreement.)

But the devil is also in the details. The reason OpenAI was able to make a deal when Anthropic could not was less about boundaries, Altman said, but about approach. “Anthropic seemed more focused on specific prohibitions in the contract, rather than citing applicable laws, which we felt comfortable with,” he wrote. 

OpenAI says one basis for its willingness to work with the Pentagon is simply an assumption that the government won’t break the law. The company, which has shared a limited excerpt of its contract, cites a number of laws and policies related to autonomous weapons and surveillance. They are as specific as a 2023 directive from the Pentagon on autonomous weapons (which does not prohibit them but issues guidelines for their design and testing) and as broad as the Fourth Amendment, which has supported protections for Americans against mass surveillance. 

However, the published excerpt “does not give OpenAI an Anthropic-style, free-standing right to prohibit otherwise-lawful government use,” wrote Jessica Tillipman, associate dean for government procurement law studies at George Washington University’s law school. It simply states that the Pentagon can’t use OpenAI’s tech to break any of those laws and policies as they’re stated today.

The whole reason Anthropic earned so many supporters in its fight—including some of OpenAI’s own employees—is that they don’t believe these rules are good enough to prevent the creation of AI-enabled autonomous weapons or mass surveillance. And an assumption that federal agencies won’t break the law is little assurance to anyone who remembers that the surveillance practices exposed by Edward Snowden had been deemed legal by internal agencies and were ruled unlawful only after drawn-out battles (not to mention the many surveillance tactics allowed under current law that AI could expand). On this front, we’ve essentially ended up back where we started: allowing the Pentagon to use its AI for any lawful use. 

OpenAI could say, as its head of national security partnerships wrote yesterday, that if you believe the government won’t follow the law, then you should also not be confident it would honor the red lines that Anthropic was proposing. But that’s not an argument against setting them. Imperfect enforcement doesn’t make constraints meaningless, and contract terms still shape behavior, oversight, and political consequences.

OpenAI claims a second line of defense. The company says it maintains control over the safety rules governing its models and will not give the military a version of its AI stripped of those safety controls. “We can embed our red lines—no mass surveillance and no directing weapons systems without human involvement—directly into model behavior,” wrote Boaz Barak, an OpenAI employee Altman deputized to speak on the issue about X. 

But the company doesn’t specify how its safety rules for the military differ from its rules for normal users. Enforcement is also never perfect, and it is especially unlikely to be when OpenAI is rolling out these protections in a classified setting for the first time and is expected to do so in just six months.

There’s another question beneath all this: Should it be down to tech companies to prohibit things that are legal but that they find morally objectionable? The government certainly viewed Anthropic’s willingness to play this role as unacceptable. On Friday evening, eight hours before the US launched strikes in Tehran, Defense Secretary Pete Hegseth issued harsh remarks on X. “Anthropic delivered a master class in arrogance and betrayal,” he wrote, and echoed President Trump’s order for the government to cease working with the AI company after Anthropic sought to keep its model Claude from being used for autonomous weapons or mass domestic surveillance. “The Department of War must have full, unrestricted access to Anthropic’s models for every LAWFUL purpose,” Hegseth wrote.

But unless OpenAI’s full contract will reveal more, it’s hard not to see the company as sitting on an ideological seesaw, promising that it does have leverage it will proudly use to do what it sees as the right thing while deferring to the law as the main backstop for what the Pentagon can do with its tech.

There are three things to be watching here. One is whether this position will be good enough for OpenAI’s most critical employees. With AI companies spending so heavily on talent, it’s possible that some at OpenAI see in Altman’s justification an unforgivable compromise.

Second, there is the scorched-earth campaign that Hegseth has promised to wage against Anthropic. Going far beyond simply canceling the government’s contract with the company, he announced that it would be classified as a supply chain risk, and that “no contractor, supplier, or partner that does business with the United States military may conduct any commercial activity with Anthropic.” There is significant debate about whether this death blow is legally possible, and Anthropic has said it will sue if the threat is pursued. OpenAI has also come out against the move.

Lastly, how will the Pentagon swap out Claude—the only AI model it actively uses in classified operations, including some in Venezuela—while it escalates strikes against Iran? Hegseth granted the agency six months to do so, during which the military will phase in OpenAI’s models as well as those from Elon Musk’s xAI.

But Claude was reportedly used in the strikes on Iran hours after the ban was issued, suggesting that a phase-out will be anything but simple. Even if the months-long feud between Anthropic and the Pentagon is over (which I doubt it is), we are now seeing the Pentagon’s AI acceleration plan put pressure on companies to relinquish lines in the sand they had once drawn, with new tensions in the Middle East as the primary testing ground.

If you have information to share about how this is unfolding, reach out to me via Signal (username: jamesodonnell.22).

The human work behind humanoid robots is being hidden

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

In January, Nvidia’s Jensen Huang, the head of the world’s most valuable company, proclaimed that we are entering the era of physical AI, when artificial intelligence will move beyond language and chatbots into physically capable machines. (He also said the same thing the year before, by the way.)

The implication—fueled by new demonstrations of humanoid robots putting away dishes or assembling cars—is that mimicking human limbs with single-purpose robot arms is the old way of automation. The new way is to replicate the way humans think, learn, and adapt while they work. The problem is that the lack of transparency about the human labor involved in training and operating such robots leaves the public both misunderstanding what robots can actually do and failing to see the strange new forms of work forming around them.

Consider how, in the AI era, robots often learn from humans who demonstrate how to do a chore. Creating this data at scale is now leading to Black Mirror–esque scenarios. A worker in Shanghai, for example, recently spent a week wearing a virtual-reality headset and an exoskeleton while opening and closing the door of a microwave hundreds of times a day to train the robot next to him, Rest of World reported. In North America, the robotics company Figure appears to be planning something similar: It announced in September it would partner with the investment firm Brookfield, which manages 100,000 residential units, to capture “massive amounts” of real-world data “across a variety of household environments.” (Figure did not respond to questions about this effort.)

Just as our words became training data for large language models, our movements are now poised to follow the same path. Except this future might leave humans with an even worse deal, and it’s already beginning. The roboticist Aaron Prather told me about recent work with a delivery company that had its workers wear movement-tracking sensors as they moved boxes; the data collected will be used to train robots. The effort to build humanoids will likely require manual laborers to act as data collectors at massive scale. “It’s going to be weird,” Prather says. “No doubts about it.” 

Or consider tele-operation. Though the endgame in robotics is a machine that can complete a task on its own, robotics companies employ people to operate their robots remotely. Neo, a $20,000 humanoid robot from the startup 1X, is set to ship to homes this year, but the company’s founder, Bernt Øivind Børnich, told me recently that he’s not committed to any prescribed level of autonomy. If a robot gets stuck, or if the customer wants it to do a tricky task, a tele-operator from the company’s headquarters in Palo Alto, California, will pilot it, looking through its cameras to iron clothes or unload the dishwasher.

This isn’t inherently harmful—1X gets customer consent before switching into tele-operation mode—but privacy as we know it will not exist in a world where tele-operators are doing chores in your house through a robot. And if home humanoids are not genuinely autonomous, the arrangement is better understood as a form of wage arbitrage that re-creates the dynamics of gig work while, for the first time, allowing physical tasks to be performed wherever labor is cheapest.

We’ve been down similar roads before. Carrying out “AI-driven” content moderation on social media platforms or assembling training data for AI companies often requires workers in low-wage countries to view disturbing content. And despite claims that AI will soon enough train on its outputs and learn on its own, even the best models require an awful lot of human feedback to work as desired.

These human workforces do not mean that AI is just vaporware. But when they remain invisible, the public consistently overestimates the machines’ actual capabilities.

That’s great for investors and hype, but it has consequences for everyone. When Tesla marketed its driver-assistance software as “Autopilot,” for example, it inflated public expectations about what the system could safely do—a distortion a Miami jury recently found contributed to a crash that killed a 22-year-old woman (Tesla was ordered to pay $240 million in damages). 

The same will be true for humanoid robots. If Huang is right, and physical AI is coming for our workplaces, homes, and public spaces, then the way we describe and scrutinize such technology matters. Yet robotics companies remain as opaque about training and tele-operation as AI firms are about their training data. If that does not change, we risk mistaking concealed human labor for machine intelligence—and seeing far more autonomy than truly exists.

Why the Moltbook frenzy was like Pokémon

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

Lots of influential people in tech last week were describing Moltbook, an online hangout populated by AI agents interacting with one another, as a glimpse into the future. It appeared to show AI systems doing useful things for the humans that created them (one person used the platform to help him negotiate a deal on a new car). Sure, it was flooded with crypto scams, and many of the posts were actually written by people, but something about it pointed to a future of helpful AI, right?

The whole experiment reminded our senior editor for AI, Will Douglas Heaven, of something far less interesting: Pokémon.

Back in 2014, someone set up a game of Pokémon in which the main character could be controlled by anyone on the internet via the streaming platform Twitch. Playing was as clunky as it sounds, but it was incredibly popular: at one point, a million people were playing the game at the same time.

“It was yet another weird online social experiment that got picked up by the mainstream media: What did this mean for the future?” Will says. “Not a lot, it turned out.”

The frenzy about Moltbook struck a similar tone to Will, and it turned out that one of the sources he spoke to had been thinking about Pokémon too. Jason Schloetzer, at the Georgetown Psaros Center for Financial Markets and Policy, saw the whole thing as a sort of Pokémon battle for AI enthusiasts, in which they created AI agents and deployed them to interact with other agents. In this light, the news that many AI agents were actually being instructed by people to say certain things that made them sound sentient or intelligent makes a whole lot more sense. 

“It’s basically a spectator sport,” he told Will, “but for language models.”

Will wrote an excellent piece about why Moltbook was not the glimpse into the future that it was said to be. Even if you are excited about a future of agentic AI, he points out, there are some key pieces that Moltbook made clear are still missing. It was a forum of chaos, but a genuinely helpful hive mind would require more coordination, shared objectives, and shared memory.

“More than anything else, I think Moltbook was the internet having fun,” Will says. “The biggest question that now leaves me with is: How far will people push AI just for the laughs?”

Read the whole story.

What we’ve been getting wrong about AI’s truth crisis

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here.

What would it take to convince you that the era of truth decay we were long warned about—where AI content dupes us, shapes our beliefs even when we catch the lie, and erodes societal trust in the process—is now here? A story I published last week pushed me over the edge. It also made me realize that the tools we were sold as a cure for this crisis are failing miserably. 

On Thursday, I reported the first confirmation that the US Department of Homeland Security, which houses immigration agencies, is using AI video generators from Google and Adobe to make content that it shares with the public. The news comes as immigration agencies have flooded social media with content to support President Trump’s mass deportation agenda—some of which appears to be made with AI (like a video about “Christmas after mass deportations”).

But I received two types of reactions from readers that may explain just as much about the epistemic crisis we’re in. 

One was from people who weren’t surprised, because on January 22 the White House had posted a digitally altered photo of a woman arrested at an ICE protest, one that made her appear hysterical and in tears. Kaelan Dorr, the White House’s deputy communications director, did not respond to questions about whether the White House altered the photo but wrote, “The memes will continue.”

The second was from readers who saw no point in reporting that DHS was using AI to edit content shared with the public, because news outlets were apparently doing the same. They pointed to the fact that the news network MS Now (formerly MSNBC) shared an image of Alex Pretti that was AI-edited and appeared to make him look more handsome, a fact that led to many viral clips this week, including one from Joe Rogan’s podcast. Fight fire with fire, in other words? A spokesperson for MS Now told Snopes that the news outlet aired the image without knowing it was edited.

There is no reason to collapse these two cases of altered content into the same category, or to read them as evidence that truth no longer matters. One involved the US government sharing a clearly altered photo with the public and declining to answer whether it was intentionally manipulated; the other involved a news outlet airing a photo it should have known was altered but taking some steps to disclose the mistake.

What these reactions reveal instead is a flaw in how we were collectively preparing for this moment. Warnings about the AI truth crisis revolved around a core thesis: that not being able to tell what is real will destroy us, so we need tools to independently verify the truth. My two grim takeaways are that these tools are failing, and that while vetting the truth remains essential, it is no longer capable on its own of producing the societal trust we were promised.

For example, there was plenty of hype in 2024 about the Content Authenticity Initiative, cofounded by Adobe and adopted by major tech companies, which would attach labels to content disclosing when it was made, by whom, and whether AI was involved. But Adobe applies automatic labels only when the content is wholly AI-generated. Otherwise the labels are opt-in on the part of the creator.

And platforms like X, where the altered arrest photo was posted, can strip content of such labels anyway (a note that the photo was altered was added by users). Platforms can also simply not choose to show the label; indeed, when Adobe launched the initiative, it noted that the Pentagon’s website for sharing official images, DVIDS, would display the labels to prove authenticity, but a review of the website today shows no such labels.

Noticing how much traction the White House’s photo got even after it was shown to be AI-altered, I was struck by the findings of a very relevant new paper published in the journal Communications Psychology. In the study, participants watched a deepfake “confession” to a crime, and the researchers found that even when they were told explicitly that the evidence was fake, participants relied on it when judging an individual’s guilt. In other words, even when people learn that the content they’re looking at is entirely fake, they remain emotionally swayed by it. 

“Transparency helps, but it isn’t enough on its own,” the disinformation expert Christopher Nehring wrote recently about the study’s findings. “We have to develop a new masterplan of what to do about deepfakes.”

AI tools to generate and edit content are getting more advanced, easier to operate, and cheaper to run—all reasons why the US government is increasingly paying to use them. We were well warned of this, but we responded by preparing for a world in which the main danger was confusion. What we’re entering instead is a world in which influence survives exposure, doubt is easily weaponized, and establishing the truth does not serve as a reset button. And the defenders of truth are already trailing way behind.

Update: This story was updated on February 2 with details about how Adobe applies its content authenticity labels.