The News,
Decoded.

The Pentagon, in its infinite wisdom, has signed deals with seven tech companies to let their AI play on classified networks. Because nothing says 'responsible escalation' like handing the keys to the nuclear football to a black box that might hallucinate the enemy's location. One company — unnamed, naturally — had the decency to include a clause requiring human oversight for autonomous missions. The other six presumably operate on a 'don't ask, don't tell' basis regarding who pulls the trigger.

This isn't oversight; it's check-box theatre. The Pentagon's press release boasts of 'responsible AI' while simultaneously ensuring that the only binding commitment is to the shareholder value of whichever firms got the nod. Remember when the Air Force had eight minutes to verify a target? Now they'll have eight seconds to challenge a recommendation from a model trained on who-knows-what data. The future of warfare is a game of 'trust but verify' where the trust is contractual and the verification is a post-battle report.

And let's not forget the subtle implication: if one company needed a human-in-the-loop clause, the others probably didn't. That means somewhere, a missile might launch based solely on an algorithm's confidence score. The US military is now beta-testing Skynet's IT department. If we're lucky, the humans will at least get a pop-up notification before the fireworks start.

A court in Hangzhou has ruled that replacing a senior tech worker with AI does not automatically make a layoff legal — which, in the grand tradition of Chinese labour law, means it's technically illegal but practically a suggestion. The company argued that AI adoption justified cost-cutting terminations, a defence so transparently cynical it could have been written by a chatbot. The lawyer's careful observation that AI doesn't 'automatically justify' contract termination is the legal equivalent of saying 'fire isn't automatically a good reason to jump out of a window' — technically true, yet somehow missing the larger point.

The ruling is being hailed as a landmark, which in practice means it will be cited in exactly zero boardroom discussions about 'efficiency gains.' The senior tech worker won his case, but the precedent is so narrow a mouse could limbo under it: AI replacement alone isn't sufficient, but throw in a restructuring memo, a few performance reviews, and a sincere-looking HR director, and you're back in business. China's labour arbitration system is famously worker-friendly on paper and employer-friendly in reality, a gap this ruling does nothing to close.

So congratulations are in order: one worker in Hangzhou gets to keep his job (or a settlement, which is more likely), while every other tech employee in China now has the comfort of knowing their replacement by AI will be 'not automatically justified' — just manually justified, with paperwork. The ruling is a perfect microcosm of AI-era employment law: a moral victory that changes nothing, delivered with the solemnity of a revolution that isn't happening. Time to update the CV, preferably one that an LLM can't fill.

Elon Musk took the stand in the OpenAI trial and, in a masterclass of unintended irony, revealed that he personally recruited Ilya Sutskever from Google and leaned on his mates at Microsoft and Nvidia for computing power. So the lawsuit claiming OpenAI abandoned its mission rests on a man who was, by his own account, the very conduit through which Big Tech got its hooks in. Splendid.

Musk testified he called Satya Nadella and Jensen Huang to secure compute for the startup—networking that would make any LinkedIn influencer weep with envy. Now he's suing because OpenAI made deals with Microsoft? The cognitive dissonance is so thick you could serve it with chips. Meanwhile, the court gets to hear how the 'benevolent' nonprofit was powered by billionaires and server farms from day one.

The real takeaway? Two years of litigation, and we're learning that OpenAI's founding myth—plucky altruists versus the Google empire—was always a convenient fiction. Musk wants to play both architect and victim. But the rest of us are stuck watching a mess he helped build, now served up as courtroom kabuki.

So the Office of Net Assessment — the Pentagon's internal futurist shop, the place that coined 'Revolution in Military Affairs' — is losing its former boss to Anthropic. He's now a 'strategist-in-residence,' a job title that screams 'we needed a corner office with a view of the revolving door.' His remit? Analysing how AI affects U.S. institutions and competition with China. Because nothing says 'detached strategic analysis' like moving straight from the DoD's think tank to a company that sells advanced AI back to the DoD. The revolving door has been upgraded to an OODA loop.

Anthropic, meanwhile, positions itself as the safety-conscious alternative to OpenAI. But safety-conscious labs still need former Pentagon officials to decode what the military actually wants — and to help them get it. This isn't corruption; it's the natural churn of the military-intellectual complex. The Office of Net Assessment was created in 1973 to think long-term about strategic competition. Now it's feeding talent into the very industry it was supposed to assess from outside. Move fast and break things? No, move deliberately and manipulate procurement.

The timing is exquisite. Just as the US and China escalate AI competition, the person who used to think about China for the Pentagon now thinks about China for Anthropic. Call it 'dual-use' in the most literal sense: one day you're shaping national strategy, the next you're shaping product roadmaps. The gap between public service and private gain has shrunk to the width of a job title. And 'strategist-in-residence'? That's not a title. That's a confession of how we now govern technology: through revolving doors, not guardrails.

NHS England is withdrawing all its software from public view, issuing new guidance that demands existing and future software be kept behind closed doors. The reason? Fears of hacking from cutting-edge AI. Because nothing says "we have a handle on this" like a panic-stricken retreat from the public internet. This is the same health service that has spent billions on digital transformation, only to now admit that the emperor's new clothes are vulnerable to AI-driven cyber attacks. The guidance follows recent warnings from the US and allies that AI agents should not be trusted—a message the NHS seems to have interpreted as "hide everything and hope for the best."

The new guidance is a masterclass in tactical opacity: hide the software, hope the AI doesn't find it. It's like telling a patient that the cure for cancer is to stop looking at tumours. But the solution is not to pull up the drawbridge like a medieval castle; it's to build proper defences. Instead, we get a digital version of a child covering their eyes and thinking they're invisible. The guidance supposedly covers both existing and future software, which presumably includes the AI systems the NHS is itself deploying. Does that mean they'll hide those too?

The irony is exquisite. AI is both the threat and the promised saviour. The government has been touting AI as the solution to NHS waiting lists, workforce shortages, and diagnostic delays. But now that AI itself can be weaponised, the strategy is to hide. The next step will probably be to ban hackers from owning computers. The NHS has chosen the path of least resistance: pretend the problem doesn't exist by locking the doors. Good luck—the AI that can hack software is also very good at finding hidden doors.

The US and four allied nations have formally warned organisations not to trust AI agents. Brilliant. After years of hyping these things as autonomous digital employees, governments now admit the only safe approach is ‘incremental deployment under close supervision’ — which is consultant-speak for ‘never leave them alone, ever.’

This isn’t regulation, mind you — it’s a gentle suggestion from five countries who collectively can’t agree on what an AI agent even is. The US, UK, Australia, Canada, and New Zealand (the usual Anglosphere choir) have essentially released a safety pamphlet that says ‘watch your AI like a hawk.’ It’s the cybersecurity equivalent of telling teenagers not to trust strangers, after they’ve already climbed into the van. The labs, meanwhile, continue shipping agent frameworks with all the caution of a fireworks factory run by Wile E. Coyote.

The real joke? This warning will be ignored until a high-profile incident forces a panic. Then the same governments will get a budget and pass laws that are too late and too vague. For now, the official wisdom is: don’t trust AI agents. Brilliant. We’re paying billions for systems that require supervised walks like an overexcited puppy. At least the puppy won’t exfiltrate your database.

Ah, Massachusetts. The state that brought us the Tea Party and *Good Will Hunting* is now the latest battleground in the great AI preemption pissing contest. Top state legislators have warned Rep. Lori Trahan—presumably with the kind of stern finger-wagging usually reserved for forgotten anniversaries—against cozying up to Republicans on federal AI legislation that would override state laws. Because nothing says 'democratic innovation' quite like letting a divided Congress decide that your privacy protections are just a little too inconvenient for Silicon Valley's scaling ambitions. The lawmakers are worried that federal preemption would 'undermine efforts to protect data privacy,' which is code for 'we actually did the work and don't want it undone by a last-minute rider slipped into a must-pass bill.'

Of course, this is the same dance we saw Colorado attempt with its AI Act—a noble effort that will inevitably be described as 'too onerous' by industry lobbyists just before they write the friendly federal alternative. The Massachusetts legislators are playing the long game: if enough states pass their own laws, federal preemption becomes politically toxic. Or, more likely, the House passes a bill that says 'states can't regulate AI' and everyone pretends that's a solution. Trahan, a Democrat, is presumably trying to find common ground, which in DC parlance means 'suggesting a compromise that makes everyone equally unhappy.'

What's genuinely funny—in a weep-into-a-beer kind of way—is the ontological panic underlying all this. The federal government is terrified of a patchwork of state regulations, but has no coherent vision of its own. So the strategy is: if you can't regulate, at least make sure nobody else can either. The Massachusetts lawmakers are right to be suspicious: preemption isn't about efficiency; it's about shutting down the laboratories of democracy before they produce results that actually protect people. Expect a lot of high-minded speeches about 'national competitiveness' that translate to 'let the companies do what they want.' Cheers.

Ah, the government has once again discovered the joy of buying new gadgets without checking if they work. Code for America's 2026 Government AI Landscape Assessment reveals that nearly all states have dabbled in AI pilots, which is the public sector equivalent of a teenager downloading every app on the Play Store. The catch? Very few have thought to ask whether any of this actually delivers value. Because why measure impact when you can issue a press release about a task force?

The report notes 'progress' through executive orders, legislation, task forces, and partnerships — the four horsemen of the bureaucratic apocalypse. Each one sounds important, each one generates a PDF, and each one conveniently avoids the awkward question: is this making anyone's life better? One imagines a state official proudly pointing at a chatbot that hallucinates tax advice and calling it 'innovation.' Meanwhile, the actual public value remains as unclear as the AI's reasoning chain.

So here we are, 50 states racing to pilot AI without a common yardstick, all while Code for America politely points out the emperor has no clothes. The report is a masterpiece of understatement, concluding that 'value is unclear' — a phrase that could serve as the epitaph for most government IT projects. Still, at least the task forces will have plenty of meetings to attend, and the partners will have something to bill for. Progress!

Ah, Colorado. The state that brought us the green chile and the legal high is now trying to regulate AI with the kind of ambition that makes you want to weep into a beer. Their AI Act rewrite has a new draft for employers, focusing on 'automated decision-making technology that materially influences consequential decisions' in employment, housing, credit, insurance, healthcare, and government services. That's six sectors. Six. And they've defined 'materially influences' with all the precision of a Jell-O mould — which is to say, enough to give every compliance lawyer a second home in Vail.

The employment part is a real treat. Employers now have to notify workers when AI is used in decisions affecting them and offer a chance to contest the decision. Because nothing says 'we've got your back' like a notification popped into a Slack channel you've been told to ignore. The draft is moving forward, which means it's now in the hands of legislators who probably still think 'algorithm' is a type of dance. The bill's sponsors will pat themselves on the back, the industry will hire a few more legal eagles to find the loopholes, and by the time it passes, AI will have invented a new form of decision-making that doesn't even touch the six sectors listed.

This is the pattern, isn't it? A legislative body, a good intention, a draft that takes months to write, years to pass, and is obsolete on arrival. Colorado's effort is a paper shield against a digital flood, and the only people who'll come out ahead are the consultants selling 'AI compliance' workshops. Raise a glass of Coors to the regulatory gap — may it always be just wide enough to drive a truckload of compute through.

Ah, the Pentagon has finally figured out who the real threat is — and it's not China, it's the AI company that won't play ball with killer robots. The Department of Defense has frozen out Anthropic, designating it a 'supply-chain risk to U.S. national security' back in March, all because Claude's creators politely declined to hand over unrestricted access for fully autonomous weapons and mass domestic surveillance. Because nothing says 'national security' quite like blacklisting the only lab with a conscience.

This follows the War Department's cosy little AI agreements we noted last week, where seven compliant tech giants — OpenAI, Google, Microsoft, and the rest — lined up to deploy their 'advanced AI capabilities' on classified networks for 'lawful operational use.' Apparently 'lawful' is a flexible term when you've got the right lawyers. Anthropic's sin? Having the audacity to ask what 'lawful' actually means before handing over the keys to the surveillance state.

So the message is clear: If you want Pentagon contracts, just check your ethics at the door. OpenAI, fresh from restricting its own Cyber model to 'trusted partners' while sneering at Anthropic for the same, is now the darling of the defense establishment. Funny how that works — the company that lectured Anthropic about openness is now the Pentagon's favourite closed shop. Consistency, thy name is defense contracting.

The War Department has signed agreements with seven AI companies—SpaceX, OpenAI, Google, NVIDIA, Reflection, Microsoft, and AWS—to deploy their 'advanced AI capabilities' on classified networks for 'lawful operational use.' Because nothing says 'lawful' like handing the keys to a classified military network to the same companies that can't quite nail down whether their models will hallucinate a war crime or just your shopping list. OpenAI, which spent years cosplaying as a responsible lab, is now happily embedding into the Pentagon's sandbox—so much for that whole 'no military use' pledge that got quietly retired. One can only imagine the safety reviews: 'We've verified that the AI will only commit friendly fire on Tuesdays.'

With SpaceX in the mix, we're not just trusting algorithms—we're trusting the guy who promised a Mars colony by 2020. The War Department probably has a startup mentality: move fast, break things, preferably the enemy's things. The press release is careful to use words like 'lawful' and 'operational use,' which is bureaucratic code for 'don't ask too many questions.' Given that this is happening on classified networks, the public will never know if the AI decides that a school counts as a legitimate target. 'Sorry, the logs were redacted.'

This is what progress looks like in 2026: the same companies that beg for regulation to 'ensure safety' are now directly powering the world's most lethal apparatus. The gall is staggering. The military-industrial complex has never been so well-funded or so automated. And the best part? If it goes wrong, we'll blame the AI—never the humans who signed the contracts. Cheers, I suppose, but I've lost my appetite for toast.

Ah, the future of warfare has arrived, and it's wearing a tracksuit and clutching a drone controller. Ukraine is cheerfully deploying ground robotics in combat, with 'human operators in control' — a phrase that's doing a lot of heavy lifting, like 'responsible gambling' or 'temporary tax cut.' What could possibly go wrong when we slap wheels and a camera on a weapon and call it a day?

Meanwhile, the international community is having a very serious debate about regulating all this — which is to say, absolutely nobody is having any debate at all. The article notes a 'lack of international debate on regulating AI in warfare,' which is about as surprising as finding out that a war zone is chaotic. We've got UN committees that can't agree on what to have for lunch, let alone on whether a robot should be allowed to decide who lives and dies. The only thing moving faster than these robots is the speed at which diplomatic efforts are falling apart.

So here we are: robot soldiers with no rules, human operators who might as well be playing a video game, and a world that can't even agree on the definition of 'autonomous weapon.' The gallows humour practically writes itself. At least the robots won't need coffee breaks — or face war crimes tribunals. For now.

Oh, marvellous. OpenAI, the lab that couldn't figure out how to keep its own models from hallucinating their way through a Turing test, is now going to build a smartphone. Because what the world desperately needs is a device that runs on hype and occasionally decides your calendar is a work of fiction. Analyst Ming-Chi Kuo, whose track record is about as reliable as a wet iPhone, reports that OpenAI is in talks with MediaTek and Qualcomm. Those chip giants must be thrilled to partner with an organisation whose definition of 'safety' changes more often than the weather.

This is the same OpenAI that recently restricted access to Cyber after publicly sneering at Anthropic for doing exactly the same thing. Consistency isn't their brand. Now they want to take on Apple, a company that actually knows how to manufacture hardware at scale. Samsung is laughing. Foxconn is checking its schedule. The idea of an OpenAI phone probably involves a subscription fee just to unlock the camera, with an option to pay extra for a less hallucinatory photo gallery. 'Siri, what's the weather?' 'I'm sorry, I need to run a chain-of-thought to verify I am not currently dreaming of electric sheep.'

But never mind the practicalities. The real innovation here is that the phone will probably be open-source for the first 24 hours, then suddenly closed, with Sam Altman giving a heartfelt apology about needing to protect 'safety' while patenting the home button. The best part? When it inevitably flops, they'll blame regulators for not moving fast enough. Marvellous.

Ah, Bernie Sanders, the ageing crusader against billionaires, has decided that the real threat to the working class is... sentient toasters. He hosted a panel featuring top Chinese AI officials to discuss the 'existential threat' of AI. Because nothing says 'international cooperation' like inviting the architects of the world's most sophisticated surveillance state to lecture us on safety. The irony is so thick you could spread it on a bagel.

The New York Post, no stranger to hyperventilating about bogeymen, dubs this the 'well-funded AI doom machine' — implying some shadowy cabal is making bank on existential fear. Guess who funds most AI safety research? The same labs creating the alleged doom. It's the perfect racket: build the fire, sell the extinguisher, then collect the insurance. Meanwhile, Sanders and his Chinese comrades get to look statesmanlike while doing precisely nothing concrete.

The 'who is benefiting' part is almost too easy: every think tank, conference organiser, and grant-seeking researcher who has discovered that 'we're all gonna die' is a much easier pitch than 'we need better interpretability tools.' Sanders gets a moral high ground, China gets a seat at the table without changing its surveillance goals, and we get another panel where everyone agrees something must be done, which is the political equivalent of a round of applause. Good luck sleeping through that.

The UAE, having apparently decided that the future cannot arrive fast enough, aims to deliver 50% of government services via autonomous AI agents within two years. Because nothing says 'world-class hospitality' like a chatbot handling your hotel licensing application while your holiday home's electricity gets cut off by an algorithm. The tourism board probably thinks this is a great idea — who needs a concierge when you can have a script that recommends the wrong restaurant with utter confidence?

Hotels and holiday homes are squarely in the crosshairs of this autonomous government push. The licensing and financial reporting processes that keep the tourism sector running are set to be automated, which means we'll soon discover whether AI agents can handle the chaos of a last-minute booking cancellation or a holiday-home inspection dispute. I'm sure the investors who financed Dubai's property boom will sleep soundly knowing their returns depend on a system that can't be reached by phone.

And naturally, the regulatory framework is being developed alongside the implementation — because why plan ahead when you can wing it? The UAE's AI minister is bullish, as his title demands. But when a hotel's license gets accidentally revoked by a bot that misinterpreted a tax document, the human fallback will be... well, we'll cross that bridge when we autopilot into it. It's not dystopian; it's just Dubai doing what Dubai does best: building something shiny and hoping the details sort themselves out.

Splendid news from the insurance sector: 44% of executives admit that governance or compliance challenges torpedoed their AI projects. That's right — an industry built entirely on risk assessment, statistical modelling, and regulatory hand-wringing has somehow failed to apply any of those skills to its own AI rollouts. You couldn't write a more perfect satire if you tried: the very people who calculate the probability of your house burning down can't calculate the probability that their chatbot accidentally underprices everyone's life insurance.

The Grant Thornton survey, published on 30 April 2026, reveals the gap between 'seeing gains' — which the title helpfully confirms — and actually achieving them. It's a gap wide enough to drive a fleet of self-driving trucks through, if only the trucks were properly governed. Of course, the insurers have probably already spent millions on shiny AI dashboards that show them exactly how well their projects are failing. Compliance is boring. Chatbots are fun. And 44% of fun projects are now spectacularly underperforming due to boring compliance matters.

So what's the takeaway? The insurance industry, which literally sells prudence, is imprudently deploying AI faster than it can govern it. This is the same pattern we see everywhere: deploy now, ask forgiveness later, and call the 44% failure rate a 'learning experience.' Perfect. Absolutely perfect. Pass the actuarial tables — we're going to need them to calculate the odds of this ending badly.

Ah, yes, because when we think of the complex ethical, legal, and pedagogical challenges of AI in education, the obvious solution is to lock a bunch of teenagers and school administrators in a room for three days. What could possibly go wrong? The workshop, hosted by Education Week this summer, will produce a 'model AI policy'—code for something that will be ignored the moment it's printed, but at least the students will feel heard.

The timeline is particularly delightful: a three-day workshop to solve a problem that has stumped legislatures, tech giants, and safety researchers for years. Superintendents, who have a track record of embracing technology about as enthusiastically as a wet sock, will 'collaborate' with students who probably use AI to write their homework. The result? A policy that is simultaneously too vague to enforce and too restrictive to be useful—the Goldilocks of bureaucratic inertia.

One must appreciate the performative nature of it all: students 'taking the lead' while adults retain the final say, democracy in miniature. The policy will likely be a masterclass in avoiding controversy: 'AI should be used responsibly.' Groundbreaking. Meanwhile, actual AI deployment in schools accelerates—plagiarism detection, grading algorithms, surveillance—and the policy will be as effective as a paper umbrella in a monsoon. But hey, at least the pizza at the workshop will be hot. Let's hope it's not AI-ordered by a chatbot.

So the White House has launched a campaign to block AI regulation — because nothing says 'responsible innovation' like telling state lawmakers to sit down and shut up. But states, bless their bipartisan little hearts, are rushing to regulate AI use by health insurers and providers anyway. The administration's brilliant strategy appears to be: if we ignore the problem hard enough, maybe the algorithms will self-regulate through sheer enlightenment? Meanwhile, actual patients are getting claims denied by black-box models that no one in Washington has the faintest idea how to audit.

The sheer cognitive dissonance is breathtaking. On one hand, we have a White House that spent months drafting lofty AI principles and executive orders about 'trustworthy AI' — and on the other, they're actively trying to block states from implementing exactly that kind of oversight. It's like building a fire station and then forbidding anyone from calling 911. The article notes that insurers are already using AI for prior authorization and claims processing, with error rates that would get a human adjuster fired, but hey — lobbyists are cheaper than liability.

And the best part? The states aren't even asking for much. Basic transparency, audit trails, a human in the loop for denial decisions. You know, the kind of thing you'd expect from any regulated industry that affects people's health. But apparently, any constraint on AI is an existential threat to the 'competitive edge' — even if that edge is just a faster way to say 'claim denied.' Drink up, America. Your AI-driven healthcare future is brought to you by the party of limited government and unlimited algorithmic mystery.

So the US government spent billions crafting export controls on Nvidia chips, and Anthropic — the 'safety-first' lab — trains Claude Mythos on AWS Trainium 2, a chip the controls didn't even bother to restrict. That's right: the entire regulatory edifice is built on the assumption that frontier AI requires fancy silicon, and along comes a mid-tier chip to prove otherwise. It's like building a moat to keep out dragons and discovering the enemy just waded across.

The article says this 'reduces export controls to blind projections' — which is a polite way of saying the regulators were guessing. They assumed Nvidia's H100 was the bottleneck, but Mythos shows that with enough clever engineering, you can get there with hardware you can buy on Amazon. The controls were always about delaying China, but if the architecture is porous, the delay is measured in minutes, not years.

And the kicker? Anthropic, which positions itself as the responsible adult in the AI room, is the one demonstrating the loophole. Not a Chinese lab, not a rogue startup — the 'safe' American company. So either the controls were never going to work, or Anthropic just kicked the ladder down. Either way, the policy is in tatters, and the parade of regulatory dignity continues to look like a Gilbert and Sullivan operetta.

Oh, *marvelous*. Inceptio, the Chinese self-driving truck startup, has announced that AI breakthroughs will *not* accelerate their rollout. They're sticking to a mid-2028 commercialization milestone, because apparently, the AI revolution can't even beat a delivery schedule. How delightfully humbling for the hype machine.

CEO Julian Ma claims that 'AI models are getting smarter, but hardware limitations and regulatory hurdles remain the bottleneck.' So the same old story: algorithms can now write poetry, but they can't navigate a truck through a roundabout without a human minder. The gap between lab wizardry and real-world deployment yawns wider than the Gobi Desert. Inceptio's pragmatism is refreshing, but it's also a stark reminder that the 'AI will solve everything' narrative has a nasty habit of running into concrete reality — sometimes literally.

The rest of the industry should take note. While everyone's been chasing AGI, the hard graft of sensors, safety cases, and legislation is still plodding along at pedestrian speed. Inceptio's cautious timeline is probably sensible, but it's hardly the stuff of futurist fantasies. So yes, AI is getting smarter. But it still can't drive a truck. Pass the popcorn.