SteamyTea

I don't know. The article is rather speculative as it is an area where not much research has been done. I think our modern life, for some, is much more 'indoors' than it used to be. How much of a real difference that makes I have no idea. I am currently sitting out in the sun. There are 13 other people here (like the last supper) and they are all indoors.

Health The shock revelation that light bulbs are wrecking your metabolism Since the 2000s, our light exposure has changed drastically as we make our buildings more energy-efficient. Now, we’re discovering how this may mess with our mitochondria, contributing to diabetes, dementia and more By Graham Lawton 10 August 2026 Red light is missing from much of our lives Emanuel Santos Take a flashlight – the one on your phone will do – turn it on and place the pad of your index finger over the bulb. See that red glow? You might think it’s something to do with blood, but you’d be wrong. It’s photons of red light bouncing around inside your finger, looking for somewhere to land. This simple experiment is a doorway to one of the most profound and surprising biological discoveries of recent years. “Light is a nutrient,” says Bob Fosbury, who was an astrophysicist at the European Southern Observatory before he turned his attention to the biology of light. That’s obviously true for plants, which use light to make carbohydrates. But it turns out it’s also true for animals and fungi, all of which exploit light in a hitherto unappreciated way. All living organisms – including us – are solar-powered. Sunlight has long been understood to be critical to life on Earth. But the fact that light is also used to oil the wheels of metabolism is only just, er, coming to light. Long wavelengths, which come from red and infrared (IR) light, are particularly beneficial for creating energy, and are readily available from sources including sunlight and incandescent bulbs. But changes in lighting intended to help save energy have inadvertently deprived us from this crucial supply. The irony is that just as we have discovered the benefit of red light, we are unintentionally starving ourselves of it. How modern bulbs starved us of red light The story begins in the early 2000s with the widespread adoption of two now-ubiquitous energy efficiency technologies. The first was the light-emitting diode (LED), which largely replaced energy-guzzling incandescent bulbs. Although this change helped conserve energy, it had other unforeseen consequences. The second was window glass that filters out IR, designed to prevent heat leaking into or out of buildings. Together, these dramatically narrowed the spectrum of light we are exposed to indoors, an environment where most humans spend up to 90 per cent of their time. “We lost about 95 per cent of the spectrum,” says neuroscientist Glen Jeffery at University College London. “Suddenly, without warning, boom, the long wavelengths go. And that’s a problem.” For the vast majority of human existence, those long wavelengths of light were unavoidable. Daylight is rich in visible red light, which has wavelengths of about 650 to 750 nanometres. It also emits vast and invisible (to us, at least) quantities of IR, which extends out to about 2500 nanometres. People who spend much of the day outside are naturally exposed to these long wavelengths, even when fully clothed. When long-wavelength light hits a human body, it sails through clothing and penetrates deep into tissues (short-wavelength light such as UV, in contrast, is absorbed by clothing, which is why covering up helps prevent sunburn). Red and IR are also abundant in moonlight and in the glow of a campfire. Indoor light, too, used to be full of these wavelengths, either from sunlight streaming through windows or light sources such as open fires, candles and oil and gas lamps. The invention of the electric light didn’t change that, as the spectrum of an incandescent light bulb closely matches that of the sun. But the adoption of energy-efficient lighting and windows robbed us of longer parts of the spectrum. Glass that filters out IR obviously blocks that portion of sunlight, and while LEDs throw out a lot of visible light, they produce nothing beyond about 750 nm. “LEDs are completely dark in the infrared,” says Fosbury. Scott Zimmerman, founder of the lighting company Silas Inc. in New Jersey who collaborates with Fosbury and Jeffery, calls LED light “ultra-processed”, an analogy with ultra-processed food. The combined effect of ultra-processed light and modern glass means that the average human today receives much less long-wavelength light than their ancestors did. Campfires, as well as candles and incandescent bulbs, emit red light ClassicStock / Alamy Stock Photo At the same time, conditions such as obesity, type 2 diabetes, dementia and early-onset cancers are on the march. The usual suspects for this are poor diet and lack of exercise. But according to Fosbury, Jeffery and others, these are only a piece of the puzzle; another reason, perhaps the principal one, is red-light deprivation. “There’s not a single cause, but I do increasingly think that light is the most fundamental and dominant one,” says Fosbury. “And it’s probably the easiest one to solve.” To understand why light is so vital, we have to go back in time again, to the Soviet Union of the late 1980s, where, at the USSR Academy of Sciences in Moscow, a biophysicist called Tiina Karu was experimenting with red and near-infrared light as a therapy. She discovered that they were useful in wound healing and pain management and wondered what the mechanism was. Eventually she hit on energy metabolism, specifically the pace at which mitochondria inside the body’s cells produce a molecule called ATP. Red light, she found, accelerated this process. Karu’s work never found a wide audience, mainly because she came from the wrong side of the Iron Curtain, according to Jeffery. But she was right, and today red-light therapy is everywhere, used to promote wound healing and treat skin conditions, hair loss, cognitive impairment, chronic pain, osteoarthritis and much more. The evidence of its efficacy is patchy, but growing, according to John Mitrofanis, a neurologist at the University of Grenoble Alpes in France. There is much debate over the correct doses and wavelengths, but researchers are in agreement that its effects are mediated by mitochondria. The key to this is the electron transport chain, a complex and highly orchestrated metabolic process that takes in energetic electrons and passes them down a chain of proteins, creating ATP in the process. ATP is biology’s principal energy currency, analogous to cash in an economy. When a cell needs energy to perform a task – say, building new proteins – it uses ATP. When mitochondria are bathed in red or IR light, the electron transport chain runs a little faster and more ATP is produced – a process Fosbury calls “photometabolism”. The exact mechanism by which ATP production is accelerated remains a point of contention, but what’s certain is that diseased cells are often short of ATP, so a boost can make up the deficit. That, says Jeffery, explains why red-light therapy is beneficial. It also explains why taking away red light can be harmful. Jeffery points out that all life on Earth evolved in an environment rich in red light and IR, but humans now mostly live in one stripped of those wavelengths. The result is more sluggish ATP production. To add insult to injury, we have also flooded our indoor environment with shorter wavelengths in the form of blue light from screens and LEDs. Blue light inhibits ATP production by slowing down the electron transport chain. Lose, lose. The effects are insidious. Fosbury likens long-wavelength light to a vitamin and calls the effects of deprivation “21st-century scurvy”, a nod to the disease of vitamin C deficiency that killed and sickened tens of thousands of sailors in the 18th century. Like scurvy, it creeps up on you. “It’s something that accumulates over a period of time,” says Jeffery. It isn’t just the general benefits of being outside, but red light specifically that is proven to have benefits. One of the symptoms of 21st-century scurvy is elevated blood sugar, which, if left unchecked, can progress to type 2 diabetes. This is due to the slowing down of the electron transport chain, which burns glucose as its primary fuel. A couple of years ago, Jeffery showed that a 15-minute burst of red light blunts the blood sugar spike after a large dose of glucose. He has also done real-world experiments in windowless, LED-illuminated offices at University College London. When he added long-wavelength light from a dimmed incandescent bulb, he found that the office occupants had lower blood sugar, on average, than their colleagues labouring under pure LED light. “Their mitochondria are being kick-started by the red light,” says Jeffery. A study published in January found something similar; researchers at Maastricht University in the Netherlands exposed 13 individuals with type 2 diabetes to 4.5 solid working days of LED light in an office environment, then 4.5 days of natural light in the same location. The participants’ blood sugar control was much better during the daylight days. Jeffery also has concerns about other diseases and health problems usually associated with ageing, especially neurodegenerative conditions. One of the causes of these, he says, is mitochondrial dysfunction, which is effectively induced by red-light deficiency. Clinical trials have found that near-infrared light can slow down age-related cognitive impairment. A recent study of almost 88,000 people in their sixties found that, over the course of eight years, those who spent more time in daylight were significantly less likely to develop dementia, though the underlying causes of this relationship aren’t yet fully understood. Mitrofanis, meanwhile, has shown that red light can prevent Parkinson’s in a primate model of the disease. Cancer and cardiovascular diseases, too, have been linked to a lack of daylight. When dermatologist Richard Weller at the University of Edinburgh, UK, and his colleagues analysed data from more than 400,000 adults in a yet-to-be-published study, they found that people with higher habitual UV exposure – a proxy for sunlight exposure in general – were less likely to die from cancer and cardiovascular disease. Indeed, says Mitrofanis, the success of red-light therapy may simply be down to correction of a red-light deficiency. “Maybe all this is doing is rescuing processes that have gone awry because we’re inside all the time, getting illuminated with blue light.” The fact that longer-wavelength light speeds up energy production isn’t a happy accident. Fosbury and Jeffery argue that it is a design feature of living systems. Life, they say, evolved to feed off this once-abundant resource to maximise the efficiency of ATP production. “Life on Earth has evolved for 4 billion years in sunlight and it’s adapted exquisitely to the properties of sunlight,” says Fosbury. This greenhouse is flooded with pink LED lights, which emit red wavelengths Getty Images/Westend61 One of those adaptations is the ability to harvest and store photons. “The body literally is designed to absorb as much as possible in the infrared,” says Zimmerman. The amount of light energy we take in every day is staggering, he says. “Sunlight is the number one energy input into the body. It’s at least twice, if not three times, as much as food.” Once sunlight enters the body, it scatters, bouncing off cells and organelles repeatedly – perhaps tens or hundreds of times, says Fosbury – until it hits a molecule that can absorb it, dumping energy in the process. This scattering means that a photon of red or IR light remains inside the body for much longer than it would if it passed straight through. This is what you see when you press your fingertip to a flashlight. The photon’s final destination is somewhat random, but occasionally it is absorbed by a component of the electron transport chain. Scientists are currently evaluating whether water molecules in the mitochondria are key to converting light into energy. When they absorb energy from a photon, they vibrate faster and lubricate the passage of electrons down the chain. Remarkably, says Fosbury, each step in the chain requires the electrons to overcome an energy barrier of roughly 0.75 electron volts, which is almost precisely what a photon of long-wavelength light delivers. Without the supply of red light we evolutionarily developed for, we’re getting far less energy. Seeking out more red light exposure Fortunately, the solution to this modern-day scourge is simple: recharge your light battery whenever you can, either with daylight or an incandescent bulb. There are red-light therapy devices, but some of these are overpowered, which can send cellular energy production into overdrive and lead to inflammation. Of course, exposing yourself to sunlight also carries its own risk. But clothing and sunscreen help block cancer-causing UV, while letting red and IR through. For people stuck in modern offices, Jeffery recommends buying a light fitting with a dimmer switch and putting it on your desk with an incandescent bulb in it – if the building managers will let you. “Even when it [the incandescent bulb] is very, very, very dim, it still produces a vast amount of IR,” he says. There’s no specific magic wavelength and so a red LED light would likely be better than no red light at all, but nowhere near as effective as full-spectrum red light from incandescent light bulbs or the sun. A London hospital has created an outdoor ward King's College Hospital NHS Foundation Trust What would really make a difference, says Jeffery, is a change in attitude among the people who design and manage modern buildings. The pursuit of energy efficiency is laudable but has unintended negative consequences. And in any case, energy efficiency need not be sacrificed. True, incandescent bulbs are extremely inefficient as a source of visible light, producing mostly infrared, aka heat. But the energy consumption of a heavily dimmed incandescent bulb is no more than an LED, so there are no extra costs – and shedloads of savings in the form of a healthier workforce. “We can do masses for public health just by bringing back the long wavelengths of light you’re being denied in your office environment,” says Jeffery. “There’s a real need for change,” says Alistair Nunn, director of science at the Guy Foundation in London, which supports research on quantum biology and its applications in medicine. “At the moment, everybody’s going down this road of ‘let’s get lots of LEDs’, but it’s not doing us any good at all.” Low on energy? A new understanding of rest could help revitalise you There is a state of relaxation that few of us spend much time in, but which comes with profound well-being benefits. With healthier ageing, reduced risk of disease and feeling more energised all on offer, here's how to get there There are signs of institutional change, with – appropriately enough – hospitals at the forefront. Studies have shown that infrared light in hospital wards leads to improved clinical outcomes, including shorter stays. And it has long been known that inpatients who have access to daylight are discharged sooner, on average, than those kept in windowless rooms. Now, some hospitals are taking action. Jeffery and his team recently did a light survey of the critical care unit at King’s College Hospital in London. “It’s a beautiful building, but there’s absolutely no long-wavelength light in there,” he says. “I said to them, ‘obviously I’m concerned about your patients, but I’m concerned about your staff too.’” The hospital managers took heed and, in May, opened a new six-bed outdoor ward on the roof. Two other London hospitals are considering outside space for their critical care patients, says Jeffery. “I think they’ll follow because they can’t afford not to.” And even if you aren’t a patient in a critical care ward, you can’t afford to risk 21st-century scurvy through a steady diet of ultra-processed light. “You can live your life without red light,” says Jeffery, “but you cannot live healthily without it.”

Down here, there is only 1 24/7 gas station. Most of the rest close at 8PM during the winter. Was not that long ago (in my mind) that the country was brought to a virtual standstill when the fuel depots where blockaded (Year 2000). Was talking to a workmate who was talking about getting a hybrid (a shit 48V one). I asked why not a full BEV. Her reply "I am going to be driving to Camborne 4 days a week". So a total of 120 miles and she can use a Granny lead at home for the 10 kWh a day it will use.

I was thinking more at the electronic/component level. Just putting a huge, variable resistor in parallel to the load would limit the current available for export, or you could quickly switch the inverter output, or even the DC input at varying rates. I am just not sure which is the most effective.

I am not sure how current gets limited.

Nearly two decades ago my first post on the Green Building Forum was about open source kit housing. I still like the idea, but it fraught with problems.

You can calibrate the RH with some salty water.

26.7°C seems quite a high temperature, is the concrete still curing? I have some very similar T/RH units, not one pair of them read the same.

Correct.

16 (amp) x 230 (nominal voltage) = 3.68 kW (power per phase)

Explain that a bit more in the context of the article.

Environment It’s good for the planet to scrap a new petrol car and buy electric Carbon savings from driving EVs are so large that it makes sense to scrap a working petrol car to make the switch Madeleine Cuff6 August 2026, updated 7 August 2026 http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wOC9TRUlfMzA3NDU2MzcxLmpwZw__.jpg Electric cars are far better for the climate than traditional petrol vehicles JUSTIN TALLIS/AFP via Getty Images In a straight choice between a new petrol car and a new electric car, on climate grounds the electric car wins easily. But what if your existing petrol car is still working fine – or is even only a year or two old? Contrary to popular belief, the overall climate impact is lower if a new petrol car is scrapped in favour of an electric vehicle (EV), new research reveals. Studies show that although EVs produce more carbon emissions when they are manufactured, they are responsible for much lower emissions when they are being driven compared with a petrol or diesel car. Overall, the whole-life emissions for an electric car are almost 90 per cent lower than for combustion engine vehicles. How I pay almost nothing to power my house and electric car Elliott Campbell at the University of California, Santa Cruz and Roland Geyer at UC Santa Barbara wanted to find out whether drivers should wait until their current combustion engine car is ready for the scrapheap before moving to electric. Using US data, they analysed the emissions impact of scrapping a fossil fuel car and replacing it with an EV at different points, from scrapping a car that is just a year or two old to one at the end of its life. Their modelling accounts for the emissions associated with producing and running a new electric car, and other factors including variability in the grid electricity mix, vehicle efficiency, manufacturing emissions and mileage. “We found that in the vast majority of cases, and even for a brand new gas vehicle, you would be far better off scrapping the gas vehicle and switching to electric,” says Campbell. “It takes more energy to build an electric vehicle, but the savings from driving the vehicle swamp all the upfront costs.” Campbell stresses he doesn’t expect people to immediately scrap their brand new cars. Instead, the study aims to show that there is zero environmental benefit in keeping a fossil fuel car on the road until the end of its life. The findings hold true for almost any scenario, except where an electricity grid is very reliant on coal power, or when a fossil fuel car is driven less than about 7000 km per year. “The idea here is just to simply show what a huge advantage electric vehicles offer, so that when the opportunity to make the transition [is] something that people are ready and excited to do, by all means go for it,” says Campbell. “There’s no environmental concern to hold you back.” Kenneth Gillingham at the Yale University says the paper takes a “clever angle”. “I think what they’re talking about makes perfect sense. I’ve thought about this before [but] I’ve never thought to write a science paper on it, and I’m impressed.” http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNS8wMS8yMzE1NDYyMS9TRUlfMjM3MDY1NDg1LmpwZz93PTkwMCZoPTYwMCZjcm9wPTE_.jpg Electric cars now last as long as petrol and diesel counterparts Records from UK vehicle safety tests show that the average lifespan of an electric vehicle is 18 years, and the reliability is still improving considerably from year to year But Gregory Keoleian at the University of Michigan points out that the study is considering an “extreme case”. “It’s unlikely somebody’s going to purchase a combustion engine vehicle and retire it after a year,” he says. Instead, those cars will likely hit the second-hand market, which could have knock-on economic and behavioural impacts, such as lowering the price of second-hand cars so dramatically that it convinces more people to switch from public transport to car use. More research is needed to model those potential impacts, says Campbell. But the study’s authors say their findings lend support to the idea of subsidised scrappage schemes, to encourage people to scrap older fossil-fuel cars in favour of electric vehicles. Such policies could accelerate the EV transition in the US and help to cut transportation emissions, which are now the single largest source of carbon emissions in the country. Journal Reference: Science DOI: 10.1126/science.adv5441 Environment It’s good for the planet to scrap a new petrol car and buy electric Carbon savings from driving EVs are so large that it makes sense to scrap a working petrol car to make the switch Madeleine Cuff6 August 2026, updated 7 August 2026 http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wOC9TRUlfMzA3NDU2MzcxLmpwZw__.jpg Electric cars are far better for the climate than traditional petrol vehicles JUSTIN TALLIS/AFP via Getty Images In a straight choice between a new petrol car and a new electric car, on climate grounds the electric car wins easily. But what if your existing petrol car is still working fine – or is even only a year or two old? Contrary to popular belief, the overall climate impact is lower if a new petrol car is scrapped in favour of an electric vehicle (EV), new research reveals. Studies show that although EVs produce more carbon emissions when they are manufactured, they are responsible for much lower emissions when they are being driven compared with a petrol or diesel car. Overall, the whole-life emissions for an electric car are almost 90 per cent lower than for combustion engine vehicles. Read more How I pay almost nothing to power my house and electric car Elliott Campbell at the University of California, Santa Cruz and Roland Geyer at UC Santa Barbara wanted to find out whether drivers should wait until their current combustion engine car is ready for the scrapheap before moving to electric. Using US data, they analysed the emissions impact of scrapping a fossil fuel car and replacing it with an EV at different points, from scrapping a car that is just a year or two old to one at the end of its life. Their modelling accounts for the emissions associated with producing and running a new electric car, and other factors including variability in the grid electricity mix, vehicle efficiency, manufacturing emissions and mileage. “We found that in the vast majority of cases, and even for a brand new gas vehicle, you would be far better off scrapping the gas vehicle and switching to electric,” says Campbell. “It takes more energy to build an electric vehicle, but the savings from driving the vehicle swamp all the upfront costs.” Campbell stresses he doesn’t expect people to immediately scrap their brand new cars. Instead, the study aims to show that there is zero environmental benefit in keeping a fossil fuel car on the road until the end of its life. The findings hold true for almost any scenario, except where an electricity grid is very reliant on coal power, or when a fossil fuel car is driven less than about 7000 km per year. “The idea here is just to simply show what a huge advantage electric vehicles offer, so that when the opportunity to make the transition [is] something that people are ready and excited to do, by all means go for it,” says Campbell. “There’s no environmental concern to hold you back.” Kenneth Gillingham at the Yale University says the paper takes a “clever angle”. “I think what they’re talking about makes perfect sense. I’ve thought about this before [but] I’ve never thought to write a science paper on it, and I’m impressed.” http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNS8wMS8yMzE1NDYyMS9TRUlfMjM3MDY1NDg1LmpwZz93PTkwMCZoPTYwMCZjcm9wPTE_.jpg Electric cars now last as long as petrol and diesel counterparts Records from UK vehicle safety tests show that the average lifespan of an electric vehicle is 18 years, and the reliability is still improving considerably from year to year But Gregory Keoleian at the University of Michigan points out that the study is considering an “extreme case”. “It’s unlikely somebody’s going to purchase a combustion engine vehicle and retire it after a year,” he says. Instead, those cars will likely hit the second-hand market, which could have knock-on economic and behavioural impacts, such as lowering the price of second-hand cars so dramatically that it convinces more people to switch from public transport to car use. More research is needed to model those potential impacts, says Campbell. But the study’s authors say their findings lend support to the idea of subsidised scrappage schemes, to encourage people to scrap older fossil-fuel cars in favour of electric vehicles. Such policies could accelerate the EV transition in the US and help to cut transportation emissions, which are now the single largest source of carbon emissions in the country. Journal Reference: Science DOI: 10.1126/science.adv5441

This is all so yesterday evening, the technology has moved on. You are using a bicycle to find a partner in the next village, the rest of us are using an app to find several all over the country. Technology Could a new approach to AI finally deliver artificial general intelligence? AI chatbots can describe reality in words, but don’t truly understand cause and effect in the real world. Now, a fresh kind of machine intelligence that does just that is emerging Daniel Cossins3 August 2026, updated 6 August 2026 http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wNy9TRUlfMzA2Njg5MTI5LmpwZw__.jpg Andrea De Santis What happens when you knock a glass off a table? Ask ChatGPT and you get a step-by-step explanation: how gravity accelerates the receptacle along a “parabolic trajectory”, why it could shatter “if the impact generates stresses greater than the material can withstand”. Ask my 3-year-old daughter and she’s less bothered with fine details: “It go smash and water everywhere!” When it comes to the future of artificial intelligence, the difference is instructive. The large language models (LLMs) powering today’s chatbots are trained on vast swathes of text to predict the next word in a sequence. Unlike little Adeline, however, no LLM has ever actually knocked a glass off a table – or thrown spaghetti at a wall, or driven a scooter into a pond – and that may impose a fundamental limitation on their capabilities. This, essentially, is why some researchers argue that what we need to take AI to the next level isn’t ever-larger LLMs, but “world models”: systems that learn through observation so they can simulate the consequences of actions in the real world. Read more A golden age of maths is dawning and mathematicians are freaking out World models have rapidly become AI’s next frontier, attracting huge buzz from industry and business thanks to their promise in autonomous robotics. What makes the world-model approach especially intriguing, though, is that it also ostensibly offers a route to artificial general intelligence (AGI), or machines with human-level reasoning that can be applied across a range of tasks. The problem is that it is tricky to parse substance from bluster. The term “world model” is confusingly elastic, for starters – for some, problematically so. It has become “a kind of shorthand for all the things that current AI systems can’t do well”, says Melanie Mitchell at the Santa Fe Institute in New Mexico. It is unclear exactly what these systems should model, never mind whether internal representations of physical reality will be sufficient for the kind of generalisable intelligence that means my 3-year-old intuitively understands what happens when she swipes a glass, even if she can’t say why. To make sense of world models as they apply in AI, it helps to understand the concept’s roots in cognitive science. AI researchers themselves typically trace it back to 1943, when psychologist Kenneth Craik wrote that the human mind “carries a small-scale model of external reality and of its own possible actions”, allowing us “to try out various alternatives” and “react to future situations before they arise”. http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wNy9TRUlfMzA1OTUxMzA0LmpwZw__.jpg Human minds have world models that enable us to predict what will happen next timsimages/Alamy That idea has evolved in the decades since, most notably with the theory of predictive processing. This posits that perception – possibly even consciousness itself – relies on the brain constantly generating predictions about the external world and updating them in response to incoming sensory data. But the point remains. “The key idea is that intelligence involves building models of the world in our heads so we can simulate outcomes and avoid costly mistakes,” says Josh Tenenbaum, a cognitive scientist and AI researcher at the Massachusetts Institute of Technology. “So ‘our ideas die in our stead’, as Craik put it.” It doesn’t take a genius to see why that has piqued the interest of AI researchers. LLMs have proved astonishingly capable, and they can certainly give the impression they understand things as we do. But language is a description of reality, rather than reality itself, and their lack of direct experience means that even the most powerful language models struggle when it comes to “understanding” physical phenomena in the real world. Read more Why I have changed my mind about AI and you should too LLMs often perform poorly when they are asked to reason about spatial concepts and things that might happen in the physical world, says Mitchell. “If you’ve ever uploaded a map to one of these models and asked it questions about it, you’ll know that they will often have a lot of problems with reasoning.” Indeed, in a 2024 study, when researchers trained language models on a database of turn-by-turn directions for taxi trips around New York City, they found that the system could provide reasonable routes from one point to another – but failed miserably when asked to take the odd detour. The reason why is that the LLMs have only descriptions of journeys, and therefore lack the sort of mental map that allows us to imagine what would happen in different scenarios. And navigation is just one example. LLMs demonstrate similar shortcomings in any scenario where you need to simulate the physical world – safely loading a dishwasher, say, or folding laundry. “They’re not designed for it, which is why people are interested in alternatives like world models,” says Tenenbaum. http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wNy9TRUlfMzA2Njg5MTgzLmpwZw__.jpg Andrea De Santis Probably the most influential proponent is Yann LeCun, a computer scientist at New York University and, until recently, chief scientist of foundational AI research at Meta. His argument, first made in a 2022 paper, is essentially that if we want to build truly intelligent systems that can reason, plan and act effectively in the real world, we need world models. “I cannot imagine we can build agentic systems without those systems having an ability to predict, in advance, what the consequences of their actions are going to be,” LeCun told artificial intelligence forum AI House Davos in January. And the key to that, he reckons, is systems that learn the rules of the world from observation. The idea has caught on. In December 2025, LeCun left Meta to found AMI Labs, raising just north of $1 billion to build world-model systems. A year earlier, Fei-Fei Li at Stanford University in California started World Labs, with $230 million of investment, to develop AI with “spatial intelligence”. Many of the established AI firms, most notably Google DeepMind, are now also actively pursuing world models in one form or another. The reason for the influx of money is primarily the promise that world models hold for advancing robotics (see “What are AI world models good for?”, below). However, it is early days and the existing prototypes are modest in their applications. With Marble, for instance, World Labs has built a system that generates coherent 3D scenes from text prompts. Similarly, DeepMind’s Genie 3 creates convincing interactive virtual environments – “snowy mountain at dusk”, say – in which you can move around for several minutes and even prompt events like rain. http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wNy9TRUlfMzA1OTUxMjc1LmpwZw__.jpg Google DeepMind’s Genie 3 creates worlds that users can explore from text-based prompts Google DeepMind In both cases, the companies describe their systems as “world models”. But Marble is really a 3D video generator and Genie 3 a video-game simulator, albeit one producing simulations in which agents could plausibly act, observe consequences and learn. Arguably, a proper world model would be something that exists inside an agent such that it can predict the consequences of its decisions, imagine the future and plan ahead before taking actions – and that is what DeepMind is pushing towards with its Dreamer 4 system, released in September 2025. It learns an internal predictive model of its environment, training mostly on data from the video game Minecraft, and uses that to train an agent that repeatedly “dreams” the consequences of possible actions to improve its behaviour. “The biggest difference [from Genie 3] is that Dreamer 4 is an agent, not just a world model,” says Danijar Hafner at DeepMind in San Francisco, who leads the Dreamer 4 team. “It predicts actions and improves them through iterative self-improvement, using planning and [imagined] trial and error.” The power of this approach is apparent in Hafner and his colleagues’ demonstration that Dreamer 4 can figure out how to collect diamonds in Minecraft, a complex task involving thousands of different actions – gathering resources, crafting tools, navigating the landscape – without being shown how to play. “The system has to understand its environment and generalise, because each new episode starts in a randomly generated world,” says Hafner. That is an important step, not least because this is precisely the kind of world model that could facilitate autonomous robots capable of folding laundry, say, or loading the dishwasher. “I think it’s going to solve robotics,” says Hafner. “What’s missing now is execution: data, compute, scaling. But we have the recipe, so, like with language models, it’s about scaling and details.” Read more The AI expert who says artificial general intelligence is nonsense Who would bet against DeepMind, given its impressive track record? After all, its researchers have won a Nobel prize for work on protein folding. But while Dreamer 4 demonstrates that agents with internal world models can reason and plan, it leaves a deeper question unresolved: what exactly should these systems learn about the world, or, more specifically, at what level of detail? And this has become a key fault line in the field, with two distinct approaches emerging. Many of the existing strategies are attempting to generate predictions on condition of action, as the researchers put it, by reconstructing future observations as faithfully to the training data as possible. But LeCun reckons that is the wrong approach, or at least not the best. His argument is based on the fact that humans don’t mentally simulate the world in any great detail, and certainly not pixel by pixel. When we imagine a glass falling from a table and smashing to pieces, for example, we don’t predict the position of every shard of glass, every water droplet. Instead, LeCun argues, we run highly compressed models that capture only the aspects that matter. This why LeCun advocates for a different tack, which he calls joint-embedding predictive architecture (JEPA). In this framework, world models infer abstract representations of what is relevant for reasoning, planning and action. “Generative models try to reconstruct pixels, whereas JEPA learns in a latent space and only predicts what is useful,” says Randall Balestriero at Brown University in Rhode Island, who has worked with LeCun on JEPA-based world models. “It ignores irrelevant details and focuses only on what matters for the agent.” LeCun and his investors appear to be betting that the JEPA approach will offer a swifter route to real-world applications, largely because it requires less training data. The most concrete publicly available demonstration of this technique so far is a system called V-JEPA 2, released by Meta in June 2025. Trained on video inputs, V-JEPA learns by masking certain regions of footage – obscuring a moving ball, say, over multiple consecutive frames – and repeatedly predicting not pixels, but abstract representations of what was hidden to learn a compact internal model of how the ball moves. What LeCun and his colleagues have shown, then, is that their system can model the causal structure of the physical world without having to reconstruct it in detail. Which isn’t to say that JEPA is necessarily any better than generative world models. Its advocates argue that it will be, of course. “Without abstraction, AI systems will stay limited to narrow tasks,” says Balestriero. But V-JEPA 2 is yet to clearly demonstrate that its abstract representations are sufficiently meaningful to enable an agent operating in an open-ended environment to reason and plan. “The biggest missing piece is the question of, how do we know that the abstraction is actually useful,” says Balestriero. “This is where there is a huge amount of active research right now, to understand: what do you capture, or how do you encode something very rich about the world that is useful for planning downstream.” Hafner, for his part, isn’t convinced that more compressed representations of reality are better. “Yann is right about many things, even if maybe he expresses them more controversially than necessary, and the JEPA approach is very promising,” he says. “But I don’t think that representations should be compressed and tiny. Ultimately, you want to learn strong representations, and what we did with Dreamer 4 [which trains with pixels but predicts in abstract space] is incredibly robust.” http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wNy9TRUlfMzA1OTUxNDI1LmpwZw__.jpg Some tech investors say that world models will be an effective way to power humanoid robots Wang Jiang/VCG via Getty Images More broadly, it is also far from clear at this stage if any of the world models in development – whether they learn and predict by reconstructing the world in high-fidelity or by inferring abstract representations – will ultimately be enough to get us to AGI. Now, it’s fair to say that AGI is another elastic term, and that claims about world models as a route towards it exist on a spectrum, with some of the most ambitious suggesting that learned simulations of physical reality could become the foundation of generally capable reasoning agents. When it released Genie 3, for instance, DeepMind insisted world models are “a key stepping stone on the path to AGI, since they make it possible to train AI agents in an unlimited curriculum of rich simulation environments”. Indeed, Hafner argues that “if you have systems that are able to represent the rules governing the world, then you’re approaching something like AGI because that’s understanding, that’s a key part of intelligence”. For his part, LeCun talks about them as a vital component of broader systems, rather than the whole story. And yet it is worth exploring the extent to which the kinds of world models in development would approximate human intelligence, because it can reveal what else might be required. Hafner says the most immediate requirement is temporal abstraction – that is, being able to reason not just about what will happen in the immediate future, but as things continue to play out. “You cannot just simulate everything at a sub-second, frame-by-frame level, because humans do not reason like that,” he says. “This is one of the open frontiers.” Tenenbaum goes further. Human world models aren’t just engines for predictions, he says, “they are much richer than that”. Human reasoning depends on various forms of causal abstraction and hypothesis-driven inference – not to mention models of other minds – which together allow us to flexibly recombine knowledge across different situations. “A key open question in all this is whether scaling up these world models will recover that richness, and my view is that it likely won’t,” says Tenenbaum. Mitchell makes a similar, if slightly broader, point. “I think the ability to have a kind of compressed, simulatable representation of aspects of the world is very important, and this notion of world models is probably going to result in useful improvements,” she says. “But I think there are probably lots of other aspects of intelligence that matter.” http://localhost:56805/wp-content/external_image/aHR0cHM6Ly93d3cubmV3c2NpZW50aXN0LmNvbS93cC1jb250ZW50L3VwbG9hZHMvMjAyNi8wMy8wNDExNTc0MS9TRUlfMjg3ODAxODkyLmpwZz93PTkwMCZoPTYwMCZjcm9wPTE_.jpg A very serious guide to buying your own humanoid robot butler You can now buy a humanoid robot housekeeper for less than the price of a second-hand car. But before splashing out, there’s something you need to know Another limitation of AI systems today, says Mitchell, is a lack of metacognition – an awareness of their own cognitive state, of what they know and don’t know, and how uncertain they are. “Is that fixable with a world model? Well, it depends what kind of world model, obviously,” she says. “But I’m a little worried that the notion of ‘world model’ is going to be used as the term for the difference between what we have now and ‘AGI’.” All of which suggests that world models as currently conceived may well be necessary for human-level machine intelligence, but not necessarily sufficient. Indeed, while it looks increasingly likely that they will be powerful and genuinely useful in robotics, and possibly scientific simulation too, it is far from a sure bet that they will replicate the way world models work in human cognition, never mind human-level intelligence more broadly. “One of the big misconceptions is that intelligence is a single thing, that there is a single world model in the brain,” says Tenenbaum. “What we actually have is the ability to run many different models depending on the context, task and goal.” Ultimately, then, the task of replicating what my 3-year-old daughter’s brain is capable of when it comes to modelling and predicting the physical world – even if it doesn’t stop her knocking glasses of water off the table – is not one to be underestimated. What are AI world models good for?

A stuck fridge or freezer compre4ssor might use that much, but should trip out. My energy monitor had the battery fail, just before it did that, it was showing 1000 times the usage

Welcome. Is your home all electric or do you also have gas? What are you doing for energy and water monitoring?

How good are you at keeping records? Could easily make your own

Been busy at work, it had been a strange year

Can you not fit the floor yourself?

First thing to look at is your time series usage. Then you know what you can easily shift (temporally), store and finally, but most importantly, change behaviour. As for PV v Batteries payback, this is a difficult way to look at it and very easy to introduce bias into the calculations. And it is kW and kWh, not kw or kwh.

Make one up. Failing that, who was the project manager and who were their insurer.

Yes, but there is a small business scheme where you pay just 12% VAT if your turnover is low. That is a (expletive deleted)ing nonsense as if you go a quid over it, you pay the full 20% and get a telling off (it is a bit more complicated than that, but we got caught a few years back)

Just had a look at the nominal value of retail sales, in May 2026, they were £9.6bn per week. If 0.01% of that did not go though the till, then that is £96m a week, or about £5bn a year (spring and summer sales are higher, so multiplied by 48 weeks). Now many people in the UK fell for BJs £350m/week on the NHS. So there is near enough a third of it just by people who think they are getting a bargain.

Then it becomes fraud. Whenever you see one of those signs in a small business 'We love Cash' often followed by reasons like 'we get to keep all of it' or 'No charges to the bank'. The trader is probably fiddling us all out of revenue. It is now cheaper to use card payments than any other method (we currently pay 1.6% per transaction value). To pay in cash at the bank costs us 3% and all the hassle of counting cash each night, checking the till float, putting right mistakes, storing a weeks takings safely and then taking it to the bank, which is realistically a 2 hour trip. Forgot to mention the 11% charge on getting change. If you run a small business these days and you do not use a card machine HMRC investigates you in a lot more detail. We had our first cashless night last month.

They could start with income tax and NI. Flat rate on every £ that passes though your hands, then the argument just becomes the rate and not the principle. They could abolish Corporation Tax, the clever companies don't pay it, it is hard to enforce, just stick it on Income Tax. Then have a Sin Tax to discourage some purchases. That could cover anything from Carbon to Crypto Currencies.

I did a quick search, would probably need a lot of unpicking to get to a £/TJ price which would be easier to understand.