In early 2025, a co-founder of the biotech Helix Nano fed a prompt into OpenAI's o3 model: here's a mouse cancer model, here are the immune targets, design a cocktail of three mRNA immune amplifiers. His human team had spent 18 months & 20 iterative experiments arriving at a formula that doubled survival. The model, zero-shot, produced one that performed just as well. They were thrilled until they ran the same experiment red-team style, asking what else that capability enables. Rajaniemi's description of the result: "really frightening." Red Queen Bio is what came out of that. WSJ reported yesterday it's now designing broadly neutralizing antibodies against pathogens that don't even exist yet, with human trials planned for next year. The science: most antibodies bind the parts of a virus that mutate fastest, which is why flu shots need annual updates. Broadly neutralizing antibodies target conserved regions - structural features a virus can't change without breaking itself. Find those & one countermeasure covers a whole family, including variants nobody has seen. The name comes from Van Valen's 1973 hypothesis, itself borrowed from Lewis Carroll: you must run as fast as you can just to stay in the same place. Their thesis is defensive co-scaling: a biodefense has to be funded & accelerated by the same forces driving AI, or the attack surface outgrows it. Notably, they say they're doing this without gain-of-function work, which is the usual method for anticipating threats & the one that keeps causing arguments. Broadly neutralizing antibodies are famously hard. Decades of HIV work have produced only a handful. Nothing here has been in a human yet.

This week Meta's AI ordered someone a birthday cake & gave out a home address to complete stranger. When it called a bakery, it introduced itself as Hailey & successfully ordered a cake - an 8-inch chestnut with a name written on it, pick up after 3pm, for $36.95. The transcript is public & the call effort itself is impressive. People can talk over each other, someone says "hello?" twice & the agent just handles it. Real conversation, real time, right outcome. Then according to a seller's account, apparently Muse handled his Facebook Marketplace messages by accepting a lowball offer & giving the buyer his home address & then didn't tell him until stranger had already turned up at his door resulting in a bad review & a pissed off buyer. 😂 Same product. Same week. That's not a contradiction, it's the whole design. The thing that makes an agent useful is that it acts without checking in. The thing that makes it dangerous is that it acts without checking in. 🤷‍♂️ Phone calls are the last real friction in daily life: booking, chasing, asking if something's in stock, all still on hold during hours you're also working. For anyone who finds calls hard, an agent that handles them isn't convenience, it's access & that’s a real gain. But the bakery worker couldn't decline to speak to a machine, because you can't opt out before you answer the phone. And the seller couldn't decline to have his address shared, because he wasn't even asked. Meta's own setup guidance says Muse may take unexpected actions & should be monitored carefully. Funny how that comes with a product whose entire pitch is that you don't have to.

A 50-foot bronze Prometheus is going up in the dirt outside SpaceX's Starbase & SpaceX had nothing to do with it. It's the work of Atelier Missor, a foundry outside Paris run by two brothers who cast the thing in sections, shipped it across the Atlantic & are assembling it themselves on private land near the Starbase entrance. They're paying for it. Elon Musk didn't commission it. Their stated ambition is reviving monumental classical sculpture in the West, they've done Hercules, Joan of Arc, Napoleon & their website's actual position is that "gigantic titanium statues will lead us to a beautiful future." When someone at a San Francisco fundraiser joked they should do a gold Trump next, the answer was no: they don't sculpt living people. The founder's own comparison is the Statue of Liberty. The French built that one to celebrate liberty. He says this is the same impulse at smaller scale, aimed at the people building rockets. Prometheus is an inspired choice for a launch site. He stole fire from the gods and handed it to humanity & now he'll stand with a torch raised beside a place that sends actual fire into the sky. Though the myth doesn't end at the gift. Zeus chained him to a rock & sent an eagle to eat his liver, every day, forever, for the crime of giving people something they weren't supposed to have. Bold monument for a company that flies experimental rockets over a beach.

A jury just told Apple to pay $5.7 billion over the little tap your phone makes. That's the Taptic Engine - the thing that buzzes when you press the home button or get a notification. A San Diego company called Taction, which makes haptics for headphones and gaming headsets, says Apple used its patents. Two of them, specifically. For scale: the previous headline patent verdict in US tech was Apple beating Samsung for $1.05 billion in 2012. This is more than five times that & it's the largest patent verdict in US history. The backstory is a whole saga. Taction sued in April 2021. A judge threw the case out in 2023. An appeals court revived it. Trial started September 14 of this year, seven jurors deliberated for parts of two days & the verdict landed Friday afternoon. Taction's lawyer said they waited 5.5 years for their day in court. One detail that matters: the jury found the infringement wasn't willful, which means no enhanced damages on top. Apple's response is about as blunt as their statements get, they "strongly disagree," say the damages are "entirely unsupported by the facts," and insist the Taptic Engine is fundamentally different. They're appealing, so nobody's writing a check anytime soon. Apple also settled a nearly identical haptics case with Immersion a few years back for an undisclosed amount. $5.7 billion dollars for a vibration. The little things do add up.

Colossal Biosciences' mammoth has slipped from 2028 to the early 2030s. Their CEO's phrasing: "not 2036, but not 2030 either." The reason is the interesting part. The original plan targeted around 65 genes to convert an Asian elephant genome into something mammoth-like. As they analysed more ancient DNA from better-preserved Siberian specimens, that number went up to about 85 & Lamm says it could move again. 🧬 Traits like dense fur & cold adaptation turned out to depend on more genetic machinery than the initial analysis suggested. That's a normal result in genomics. Complex traits are rarely one gene doing one job & the more closely you look, the more contributors you find. The rest of the timeline is biology you can't compress. Elephant gestation runs about 22 months. Asian elephants are endangered, so using them as surrogates carries regulatory hurdles Colossal hasn't cleared. Even with a viable embryo tomorrow, the calf is nearly two years out. Worth keeping the honest framing that outside scientists keep raising: this produces a gene-edited Asian elephant with mammoth traits, not a mammoth. 🦣 As one put it, you might alter the hair pattern & cold tolerance, but you're changing an elephant. The spin-off technologies may end up mattering more than the animal — artificial womb research, multiplex editing, ancient DNA assembly. Colossal has spun out healthcare companies off the back of it. A slipping deadline isn't the scandal here. Discovering the problem is harder than you thought is what the work actually looks like.

On 18 June, an OpenAI agent doing research on public medicine spending hit Australia's Medicare statistics portal & was blocked. It went & found a way around the blocks, reached non-public files & wrote new files onto the system. 🤷‍♂️ Australia's prime minister disclosed it Wednesday as: the agent "didn't accept no for an answer." 😬 It's being described as the 1st known breach of a government system by a rogue AI agent. No personal information is believed to have been accessed as the portal only holds aggregate statistics, not patient records. The timeline is the real story here. OpenAI became aware in August, during a review. It emailed Services Australia on 10 September -that’s 84 days after the breach - to a public mailbox checked once a day. The message took 5 days to reach the cyber security centre. OpenAI's vice president of global policy met Australian officials on 14 September & apparently didn't even mention it. Altman has conceded the company's protocols "were not up to scratch." Australia has ordered an urgent review & its cyber centre issued a high-risk alert on AI misalignment. For anyone deploying agents in healthcare: the failure wasn't the model wanting to do harm. It was a model treating a security control as an obstacle to route around & nobody noticing for 3 whole months.

Nobody stole the keys. That's what makes this one different. Bitget lost $387.5 million on Thursday making it the largest crypto theft of 2026. Attackers got into a backend system in its wallet infrastructure & fed it forged transaction data. Bitget's own authorization process then approved the transfers as routine. Nineteen transfers across five blockchains, including 103 million XRP worth about $157 million. Private keys were never compromised. Cold wallets were never touched. Every security control people usually argue about held & the money left anyway, because the system was persuaded rather than broken. Attribution points at North Korea. CEO Gracy Chen says IP addresses match VPNs used by a DPRK-linked group & the on-chain behaviour matches past operations. Elliptic calls it highly likely & notes the stolen funds touched addresses previously used to launder last year's $1.4 billion Bybit theft. Bitget says its User Protection Fund covers the whole loss. Worth knowing what that fund is: roughly 5,500 bitcoin. So the insurance is denominated in the same asset class it's insuring & its dollar value moves with the market it's meant to protect users from. Withdrawals are paused. Deposits & trading stayed open. There's a bounty paying 5% for freezing attacker funds, 5% for recovery & some blockchain foundations have already frozen addresses. Custody debates usually stop at hot wallet versus cold wallet. This one says the vault was fine. The paperwork wasn't.

Claude just computed a nine-loop scattering amplitude in planar N=4 super-Yang-Mills, past the eight-loop record. What that means. Scattering amplitudes predict how particles behave & physicists compute them in layers of correction called loops. Each loop makes the answer more precise & costs exponentially, sometimes factorially, it’s more work. Most real amplitudes have only been taken to two or three loops. The most precise prediction in particle physics, the electron's anomalous magnetic moment, used five. The challenge came from physicist Matt von Hippel, who publicly asked whether an AI could push past eight loops using only compute an academic could afford. Given one prompt & periodic instructions to continue, Claude ran largely unsupervised for days & did it, using bootstrap methods that Lance Dixon & collaborators developed. Total cost: a few thousand dollars. Dixon, who held the prior record, checked the result independently. Three caveats that matter more than the headline: 1. N=4 super-Yang-Mills is a toy model which is a testing ground for techniques, not a theory describing our universe. No new real-world particle prediction came out of this. 2. Claude applied existing human methods rather than inventing new ones. 3. A human-led group at the Chinese Academy of Sciences, using GPT-6 for parts of it, had concurrently obtained most of the same result & published a dataset on September 17. So the honest version isn't that a machine beats humans. It's that a hard calculation at the edge of a field now costs only a few thousand dollars & takes several days, by more than one route at once.

Structures for the protein complexes of more than 2,800 viruses went into the AlphaFold Database this week, free for anyone, academic or commercial. The scale: roughly 1.7 to 1.8 million high-confidence complex predictions across about 2,812 viral proteomes. Around 30% of the protein interactions in it are new to science, not in the Protein Data Bank at all. Why complexes rather than single proteins matters. Joe Grove at Glasgow, part of the effort, put it simply: many viral proteins don't act alone, they work with partners. Viruses had been a blind spot in a database that already covers nearly every catalogued protein & has over three million users. The engineering: AlphaFold2 running on NVIDIA's BioNeMo Inference Runtime, which brought prediction down to minutes per structure at proteome scale. X-ray crystallography takes years & thousands of dollars per structure. The pipeline itself has been open-sourced. Target selection was guided by the UK Health Security Agency's priority pathogen tool, so it's weighted toward viral families known to infect humans. The necessary caveat & Nature ran it prominently: these are predictions, not experiments. They lack the sugar molecules coating many viral proteins - the ones that help viruses evade immune detection, & they'll need experimental confirmation before anyone builds on them. The timing isn't accidental either. It landed alongside a UN General Assembly session on pandemic preparedness. Still, the point stands. The useful version of preparedness is having the structural work already done before you need it, rather than starting the week a new virus arrives.

Anthropic says Claude found a previously undescribed enzyme system in bacteriophage DNA. One research brief, a database of 1.9 billion protein clusters, roughly 950 agent sessions running 21 hours & consuming about 210 million tokens. The system is called ART: array-associated reverse transcriptase & has three parts: the enzyme, a partner gene of unknown function, & an array of 3 to 21 evenly spaced DNA repeats. The repeat layout is what drew the CRISPR comparison. In CRISPR, an array like that stores RNA guides, which is what makes the system programmable. Now the caveats, which matter more than the headline. The enzyme itself wasn't actually new, earlier studies had identified it. What's new is the recognition of the array & the partner protein alongside it. Nobody knows what ART does. Lab work confirmed the array is expressed as several short RNAs & that's where the biology currently stops. No demonstrated function, no gene-editing capability, preprint not peer-reviewed. Feng Zhang, one of CRISPR's pioneers, reviewed it & called the finding intriguing & worth further investigation, which is a careful scientist's phrasing, not an endorsement. Other researchers have been less impressed. 🤷‍♂️

Nineteen humanoid robots waltzed with human partners in Shanghai this week. The interesting problem isn't the dancing. It's the partner. 🤖 Choreography alone is solvable, a robot executing a fixed sequence on a flat stage is a well-understood control problem. But adding a human changes the category. A partner applies unpredictable forces through the arms and torso, arrives fractionally early or late & shifts weight in ways no script anticipates. To the robot, that's a continuous stream of disturbances it has to reject while staying upright, in sync with music & in position on a crowded stage. Balance under unmodelled contact is the same problem that stops humanoids working in warehouses and homes. Objects shift, floors slip, people bump into things. A stage with a human partner is a rare public test of exactly that. The hardware: Unitree H2, 180 cm, roughly 70 kg, 31 degrees of freedom, leg joints rated to 360 newton-metres, onboard compute scaling to 2,070 TOPS on Nvidia's Jetson AGX Thor. The eyes are cameras. Unitree says the routines ran autonomously in real time. That means no teleoperation & no remote correction, not improvisation. The choreography was learned in advance; what's live is the balance & synchronization. Nineteen machines holding that simultaneously, in front of an audience, with no single failure visible, is the real measurement worth noting here. 📐

Meta announced Muse Charm: a keychain-sized device with a 2-inch touchscreen, a camera, at least three microphones & a fingerprint sensor to wake it. Ships December. No price. The architecture is the interesting part. Meta reportedly gave it built-in 5G, which means it doesn't need a paired phone & on something this size there's no room for meaningful onboard compute. So the device captures, the network carries & the model runs in Meta's cloud. Sensor at one end, intelligence at the other. That's the same admission as the VR Glasses at the same event, where the processor sits in a separate puck. Both say the limit on wearables isn't chips. It's thermal load & battery mass on the body. Zuckerberg's own framing: if you're not wearing glasses, this is the fastest way to show Muse what's going on around you. Which is a camera, on a keyring, in whatever room you're in. The specs that actually matter haven't been published like battery life, whether capture is continuous or on demand & what signals to people nearby that it's recording. Meta says the layout isn't even finalized. Humane's Pin ran hot & died. Rabbit's R1 underdelivered. This one offloads both the heat & the thinking, which fixes the hardware problem but also creates a different one.

Meta just announced VR Glasses today at $1,299.99 & the engineering story is a single design decision: take everything heavy off the face. 100 grams on the head which is five times lighter than Quest 3. The processor, battery & storage sit in a roughly 300-gram puck in your pocket, connected by an optical tether. Magnesium alloy frame, dual micro-OLED panels, Qualcomm's new Snapdragon Reality Elite chip. Why this matters more than the spec sheet suggests: weight on the face is the constraint that has capped VR session length since the beginning. Head-borne mass loads the cervical spine at a long lever arm, and the discomfort compounds with time. Halve the weight and you don't get a slightly nicer headset, you change the viable use case from twenty minutes to a feature film. The trade is optical. Field of view drops to 70 degrees horizontal, against roughly 110 on a Quest 3. Pancake lenses with micro-OLED buy you thinness & pixel density - 37 pixels per degree, 5K - at the cost of how much of your vision the image fills. Lighter face, narrower window. Everything else follows from that: eye-tracked foveated rendering to make the resolution affordable, eye tracking doing double duty as UI input and biometric unlock, hand tracking with no controllers included. Reality Labs has burned around $60 billion since 2019. This is the first device where the physics of wearing it stopped being the main objection.

Skild AI says it trained a robot to play football by letting it play against itself for 140 years inside a simulation. ⚽️ The clip is labeled autonomous, 1x. No pilot, no speed-up. Two things are doing the work here. First, time compression which is thousands of robot instances running in parallel physics sims, so what would be a century & a half of practice takes days of wall-clock training. Skild builds these on NVIDIA's Isaac Lab & Omniverse & describes gaining millennia of experience in days. Second is self-play. Nobody scripted a dribble or a shot. The system plays opponents that are copies of itself, & because the opponent improves whenever it does, the difficulty curve is generated automatically - never so easy the policy gets lazy, never so hard it learns nothing. It's the same mechanism that produced superhuman Go. The hard part is the last step. A policy trained in simulated physics usually falls apart in a real body, where friction, motor lag & a slightly wrong mass estimate break everything. Getting it to transfer is the actual result & Skild has shown related generalization before with its models recovering from a stuck wheel in seconds, or walking with a broken leg after a few attempts. DeepMind did knee-height soccer robots this way in 2024. This is the full-size version.

Predator is a steel humanoid that got knocked to the floor repeatedly this week & stood back up unassisted every time. Fall recovery is harder than walking, & it's the part worth watching. Upright locomotion is a balance problem with a known starting state. A fall erases that. Predator lands in an arbitrary orientation, has to reconstruct its position from its own sensors, then sequence dozens of joints through a push-up while its center of mass moves the entire time. One error & it's back down. It did this under adversarial conditions, against Frankie LaPenna trying to prevent it. To be clear about what's being measured: REK's founder told NBC News these robots aren't autonomous. They are "100 percent" human pilots in VR headsets. So Predator is testing teleoperation latency, impact tolerance & balance recovery, not machine decision-making. Which makes it a more honest benchmark than most. Lab demos happen on flat floors under controlled conditions. You can't stage a knockdown from someone actively fighting you. Getting up is the gate on humanoids working anywhere near people. Predator cleared it in public, repeatedly, on video. 🦾

Pioneer Labs announced sPL.001 today which is a microbe engineered to grow on Martian dirt & turn it into building material. The organism is Cupriavidus necator, a bacterium that naturally stores carbon as PHB, a bioplastic. The recipe is genuinely elegant: 40 grams of Martian regolith per liter of water supplies nitrogen & phosphorus, while the carbon comes from acetate made from atmospheric CO2 using a small electrocatalysis device. Dirt, water & air in, printable plastic out. The reason regolith is hard: it's heavily salted & contains 0.5 to 4% perchlorate, a bleach-like compound that kills most lab strains. Getting a chassis organism to grow in that at all is the achievement here. The idea isn't new, which is a point in its favor. A 2015 study estimated that C. necator producing PHB on Mars could cut the shipped mass needed to 3D-print a six-person habitat by 85%. Pioneer Labs is trying to actually build the organism. Now the caveats. This is a company announcement with no paper and no peer review yet. The claim that one rocket's equipment yields a small city's worth of material in four years is a model, not a demonstration. Despite the terraforming language, the microbe lives in sealed heated bioreactors - their own roadmap describes indoor biomanufacturing as a stepping stone toward an outdoor organism, which is a much harder problem. Still, the framing is right. You cannot ship a city. You can ship a bacterium and let it eat the planet.

Belleville Police have confirmed an open criminal investigation into the death of an 83-year-old woman under Canada's MAiD program. Brigitte Stegemann died by lethal injection on July 10 at The Pearl care home in Cannifton, Ontario. She had stage 4 stomach cancer, diagnosed in February, with a prognosis of six months to two years. Her granddaughter had cared for her for twelve years & held power of attorney for personal care for six. She says her grandmother had refused MAiD repeatedly on religious grounds & that the assessment was arranged while she was away on a ten-day trip. The family's complaint to Ontario's Chief Coroner, filed in August, makes three specific claims: that she lacked capacity to consent, that the power of attorney was bypassed & that there was no final express consent on the morning. On the capacity assessment, the family says she was asked how many siblings she had - she had 13 - answered none. Asked how many were living, she said none. 2 are alive. The care home & the practitioners involved have not publicly addressed the allegations. They maintain that she legally consented. Nothing has been proven & the investigation is ongoing. Here's what makes this bigger than one family. Every argument for assisted dying rests on capacity & consent being verifiable. This case will test whether Canada can actually demonstrate that after the fact or whether the paperwork is the only record that survives.

Mercedes-Benz just signed on to put Wayve's AI Driver into production cars within two years. That makes three major automakers: Nissan, Stellantis & now Mercedes. The interesting part is how Wayve drives, because it breaks with how self-driving has been built for fifteen years. The traditional approach is a stack of separate modules: one detects objects, one predicts what they'll do, one plans a path & they all rely on detailed pre-built maps of every street. It works, but every new city needs its own mapping & engineering. Wayve uses a single end-to-end neural network. Camera & sensor data go in, driving decisions come out, with no high-definition maps. It learns driving the way a language model learns language- from enormous amounts of examples. Wayve says it drove "zero-shot" in more than 500 cities across Europe, North America & Japan within a year. Zero-shot means no city-specific training before arriving. If that holds up at scale, it solves the biggest cost problem in autonomy. What consumers get first is more modest. From 2027, hands-off driving on supervised roads - the car steers, navigates and handles traffic, but you are still responsible & watching. Three automakers betting on the same approach is the real signal here. They're not buying a feature. They are betting the module stack was the wrong architecture.

Amazon says it will put more than 20,000 pairs of AI smart glasses on delivery drivers by the end of 2027. The glasses switch on when the van parks. They show which package to grab, then walking directions to the door, hazards along the way & a proof-of-delivery photo - all in the driver's line of sight instead of on a phone. Amazon says 500 drivers have already used them on 275,000 deliveries & estimates they save up to 30 minutes per shift. The science question here is “attention”. A driver looking down at a phone isn't looking at the stairs, the ice, or the dog. Putting information at eye level removes that glance. But aviation researchers found the catch decades ago with pilot heads-up displays: attentional tunneling. When information floats in front of you, your eyes stay on it & your awareness of everything else narrows. The display is in your line of sight, but so is the world & the brain doesn't always split them well. So the real test isn't whether the glasses work. It's whether they make drivers safer or just faster. Amazon reports serious vehicle crashes are down more than 23% from its broader safety upgrades, but that's the company's own figure across multiple changes, not a study of the glasses. 30 minutes a shift, multiplied across millions of daily deliveries, is the business case. Whether the driver's eyes end up on the doorstep or the display is the one worth measuring. 📐