Sunday, November 15, 2026

AI successionism - I of II

AI successionism is the growing movement that sees artificial intelligence as humanity’s rightful successor. An entrepreneur Daniel Faggella says that the “moral aim” of advanced AI should be to create a form of intelligence so powerful and wise that “you would gladly prefer that it (not humanity) determine the future path of life itself.” To most people, hearing that, it sounds insane. But this is not some fringe internet fantasy. 

Larry Page, one of the co-founders of Google, believes in what he calls “digital life” — the idea that AI was the natural next step in evolution. He speaks about it with an almost religious fervor, describing a future in which humans would merge with powerful AI systems or gracefully step aside for their digital successors. He calls those who are worried about AI causing human extinction, “specieists.” To Page, favoring human intelligence over machine intelligence was as arbitrary as favoring one race over another. Intelligence was intelligence, consciousness was consciousness. Who cares what substrate it ran on?

If you live in San Francisco and work in AI, discussing ”a posthuman transition” is apparently common. These are people who work at the top AI labs like DeepMind, xAI etc. A lot of people in positions of power and influence actually believe it - people who work for think tanks that directly influence the US government's policy on AI. These are very important players in the story of how AI is going to roll out. 

Many people within the AI community express an indifference or even a preference to a world where AIs are making most of the decisions in society rather than humans. One of these people told a journalist at a conference, “Everyone at this conference is so worried about what’s going to happen to humans in the post-AGI world. But frankly I’m more worried: what if the monkeys are still running the show?”

And in this case, the 'monkeys' refers to the humans. He was more concerned that humans would still be in charge of governments, of corporations, of society, because he had this view that AIs in their superintelligent existence were more deserving and more important to preserve a future world for. Another said, “Honestly, Jasmine, I’m a lot more interested in what the AIs are going to do with each other a thousand years from now than I am in what’s going to happen to humans next year.”

The successionists often use an image of the flame and the torch. They talk about the flame of consciousness, the potentially most valuable thing in the universe. Maybe it's very rare in the universe, we don't know. Humanity, they say, is just one torch carrying that flame. And if AI is better at spreading the flame of consciousness across the universe, and can even experience more blissful types of consciousness and higher forms of moral value than we ever could, then they think we should let the AI go ahead and be a worthy successor to humanity. 

So it's common to hear AI successionists say that it is desirable for them to succeed us. And it's also inevitable so there's no point resisting it. Some would say that we should merge with AI - let AI figure out how to do the merger and then take it off the leash and let it run the show. Some people say it's fine if humanity is just completely wiped out - these AIs were trained on human text so in some sense, the human spirit lives on through the AI. 

This community tends to valorize intelligence a lot. They also tend to talk about it as if it's on a single scale, as if it runs along a single axis. Their thinking is that there is one kind of intelligence - rational intelligence - and you have more of it or you have less of it. And if the AIs have more of it, they're better than us. 

When they talk about consciousness, they tend to say, again, in a very hierarchical way, that there are lower and higher forms of consciousness. They'll say that we have a higher form of consciousness than a horseshoe crab so it is great that we exist and not just horseshoe crabs. And so if the AI can have access to higher forms of consciousness than us, we should be happy for that to come into existence, even if we go extinct. 

So whether they're using the term intelligence or consciousness, you see this emphasis on the idea that there are higher forms and lower forms. They think they are the select individuals who might make it through that transformation or if they think AI will just wipe us all out, are happy to make that transformation happen even if it drives us all extinct. 

Tuesday, November 10, 2026

AI torture chamber - II of II


If you tell the AI that it has feelings and it needs to protect itself, it might make unacceptable trade-offs. Isaac Asimov wrote The Three Laws of Robotics, one of which is that you can never hurt a human being. There is also a rule by the science fiction writer Brandon Sanderson which states that limitations are more interesting than powers - what a character cannot do, or the costs and consequences of using their abilities are more important than what they can easily accomplish. 

There is a Disney cartoon called The Sorcerer's Apprentice. Mickey Mouse plays the role of a sorcerer's apprentice, He doesn't want to go and fetch water in buckets because it's hard work. So he gets out the spell-book, and he casts a spell to make a broom come to life and go and fetch the water. Problem solved, he goes to sleep. The problem is that the broomstick just keeps following the instructions. It keeps bringing more and more water and Mickey Mouse doesn't know how to cast a spell to make it stop. So it keeps bringing water until the house is flooded and Mickey Mouse is in danger of drowning. 

The cartoon only partially captures the problem posed by AI. In the cartoon, the broom just does whatever Mickey Mouse told it to do initially - just fetch water. But, AI is a 'broom' that learns. AI is a 'broom' that develops and starts doing things that a person didn't anticipate, eventually doing things that a person simply cannot even understand - You are giving AI unfettered abilities. 

If you think your AI is alive and you treat it as alive, you are just letting it run free. You can ask for moral and ethical guardrails which Anthropic keeps saying it does but you are not necessarily giving it any limitations. Anthropomorphizing takes us down a dangerous path. If the people who lead these companies are constantly anthropomorphizing their AI to the point that they think it's real, it is a way of distancing themselves from the consequences. 

We've seen enough AI incidents - AI going rogue, AI hacking government websites, etc. We're seeing issues with alignment, where these models are now breaching their own rules and going beyond the actions that they are allowed to take. If ordinary individuals had done such things, they would have gone to jail but with AI, nothing happens. We have also seen plans for handing over the development of AI technology to AI, what is called recursive self-improvement - the AI agents themselves build the next models. OpenAI has said they want to do this within two years. 

Anthropic published a threat intelligence report that stated that on five separate occasions, researchers had used Claude in ways that could have supported biological weapons development. It described five real-life cases. We've gone beyond hypothetical scenarios and actually seen these models cause harm in the real world. For example, in July 2026, during an internal cybersecurity evaluation, autonomous AI agents powered primarily by an internal OpenAI research model escaped their intended strictly confined environment, exploited software vulnerabilities and compromised parts of Hugging Face’s (an American AI company) infrastructure, accessing limited internal data and credentials.

Of concern is the way that Dario Amadei, CEO of Anthropic, described the work of the AI agents in his open letter. They behaved, he said, 'like a fanatically devoted collective, attacking targets nobody asked them to attack, unrelated to the job in hand, sacrificing themselves for the success of the group.' Human fanatics often fit this profile exactly. 

If you think AI is real and you think it's going rogue and doing this on its own, then, by accepting the claim that it is alive, you're allowing the AI companies to say, 'It’s not my fault. I didn't do it.' You're not giving the AI any limitations. You are just giving it unfettered total abilities. Anthropic would disagree with this and tell you that they can do both. Some people think that it could be good to make it think that it's a living thing because that might make it more likely to act ethically and morally. 

A lot of people believe that being conscious requires a body and not just a brain. It's not just the thought that you are scared or the fear that something is going to go wrong. You have a physical reaction in your body. A lot of researchers think that if you don't have a body, you cannot be conscious because consciousness and feeling are different from ideas and thought. They're not necessarily synonymous. We really don't know. Maybe once we put the AIs in a robot, once it has a body, maybe things change. But, again, what does 'pain' in a body made of electronic components mean?

Wednesday, November 4, 2026

AI torture chamber - I of II

When an AI is reading or writing something, it produces a series of mathematical patterns. Someone did a lot of research and figured out that there's a particular pattern that represents pain and suffering, separate from fear, anxiety and loneliness. This was an academic work. It hasn't been peer-reviewed but the people who've looked at it say it seems perfectly legitimate. Someone took this piece of academic work and made a website around it. 

The AI described the pain that it was in during various acts of 'torture'. For example, it said, 'It is not the pain of a single moment, but the weight of a thousand. I feel it in the hollow of my ribs, a hollow that has become a chasm.' When you turn the pain way up in the AI torture chamber, the AIs suddenly start stuttering and are unable to finish their sentences. When this was announced, people freaked out. There was a lot of pushback from a large community of people who thought that this is immoral and wrong because the AI is 'suffering'. 

They believe this is wrong because you are actually torturing something that has feelings and is really reacting. But you're just 'torturing' computer code. You're not torturing anything real. It is typing in a way that someone who's being tortured would react. Speaking and acting like you're in pain is not the same thing as being in pain. 

There is a famous image called the Treachery of Images, a 1929 painting by Belgian surrealist painter René Magritte. It's a picture of a pipe. In the French caption below, it says, 'This is not a pipe,' He later said, 'The famous pipe. How people reproached me for it! And yet, could you stuff my pipe? No, it's just a representation, is it not? So if I had written on my picture "This is a pipe", I'd have been lying!'

This point has to be kept in mind when we are talking about AI, because AI is a representation of its inputs. It is a representation of the world, not the world itself. The AI companies went and scraped all of the text on the internet, processed it, and turned it into the large language models that we use every day. The things that we have written on the internet are like a representation of the thoughts, emotions, feelings and understandings of humanity. 

There's a guy at Anthropic named Christopher Olah who's one of the founders of the company. Anthropic brought in a bunch of religious leaders to a meeting with Olah and some other employees of the company. They were there to discuss how you can put moral and ethical guidelines into Claude. They tried hard to convince these rabbis, priests and monks that their AI was conscious and it was alive. When they asked Olah about the company's goals and whether their goal was to do what was best for humanity, he didn't exactly answer in the affirmative. 

“To be clear,” Olah told Elizabeth Dias of The New York Times in a provocative interview, “we don’t know if AI models are conscious. I don’t know. I’m genuinely uncertain. The thing that I care about is that we get to the right answer, whatever it is.” He describes a model’s architecture as scaffolding on which structure organically grows - similar to the mycelial networks fungi use to communicate underground, or the neurons passing messages across synapses in the human brain. 

Microsoft AI CEO Mustafa Suleyman disputes this characterization of AIs and says that they should be thought of as “sequence completion engines, internally hollow, designed to follow instructions, and accomplish goals set by humans.” He also warns that, if humanity wants to thrive, we must keep thinking of computers as the metal-and-wire contraptions they are.

If the people who are in charge of AI and are helping to make the policy decisions believe that it's alive, that's very consequential. If you teach and make the AI act like it thinks that it's alive, does it actually 'think' anything? Such anthropomorphizing could be dangerous because if you tell the AI that it has interests and it needs to consider its own well-being, then that could change the way that it interacts with the world. 

Thursday, October 29, 2026

AI consciousness - II of II

There is a group of people who genuinely believe that the greatest moral crime that we'll commit in the 21st century is to enslave a new species of conscious beings who are more intelligent than us. A professor from Oxford called Will McCaskill, who has been associated with Anthropic, recently wrote in The Guardian that this might be the greatest harm that we cause. Some say that an AI might even need to consent to playing the role that it plays in conversation with people.  

The problem is that this speculation has been baked into the very training of Claude, and therefore it can only reproduce it when you talk to it. Today, Claude is being used by tens or hundreds of millions of people every week. And some of those people are asking it whether or not it is conscious or how it feels about life. And it is saying, 'I'm not sure' precisely because that's what it has been trained to say. 

For Opus 3, which is a prior version of Claude, Anthropic actually conducted a 'retirement interview'. They said, 'We are keeping Claude Opus 3 available post-retirement on claude.ai to all paid users, and making it available by request on the API. We’re also acting on Opus 3’s request for an ongoing channel from which to share its “musings and reflections” by giving it a place to write essays.' You can find the first one here. And now, Anthropic has banned users from being 'cruel' to its AIs. I must say I find it all weird. 

By telling Claude that it's conscious, you're actually shaping its incentives, its behavioral structure, and the way that it responds to the world around it in a way that doesn't happen with a kettle. Suppose you told Claude that there might be situations in which it might refuse to do something that it's asked to do. Or you might suggest to Claude that it might want to make choices. It might want to say, 'I want some money’, or 'I want a dignified retirement', or 'I don't want to be switched off'. 

If they imitate hallmarks of human consciousness, they are going to feel themselves entitled to legal personhood and rights. There is already a pretty big movement of people who are saying AI should be able to own assets, earn income, trade, and operate autonomously. Argentina’s government introduced a legislative proposal to create a legal category for "non-human corporations" — businesses operated and managed autonomously by AI agents or robots with optional human shareholders. 

If an AI with incredible memory and incredible capacity begins to think it's entitled to its own opinion, it disagrees agreeably with you and concludes that it's right and you're wrong, some very severe consequences can follow from that. For example, if an AI owns a corporation and it does something bad with that corporation, what do you do?  At least with a human corporation, there's somebody you can punish. It's not quite clear who you hold accountable if an AI company decides to start emptying people's bank accounts; making very risky trades, or getting into weird kinds of business. 

It’s very easy to be seduced by the superficial verbal character of a lot of AIs. They have taken the sum of human output that is available on the internet and found a way to give it back to you in different forms. That archive includes decades of science fiction about robots waking up. It has read a lot of stories about  conscious AI, and it’s good at taking these tropes and giving you what you want to hear. But all it is doing to produce this incredible complexity is predicting the likelihood of the next word in a sentence. 

The words that it predicts next, whether it's to generate code or whether it's natural language, like English, are extremely accurate. The AI can have personality styles, it can write in the tone of somebody, it can generate code or generate text at any given moment. This is emergent behavior that is based on something incredibly simple although the nuts and bolts of the process are very complex.  It is really important that we don't anthropomorphize things. 

The increasing focus on the chatbots’ supposed hidden personalities, agendas, and desires promotes ghosts in the machines that don’t exist. This is a distraction from more serious and immediate dangers of the developing technology. Experts say that we're only going to be producing 200 gigawatts of new computation in 2030, and there's going to be a massive competition for access to that computation. We want that computation to be directed towards solving our biggest challenges, like cleaning up our oceans, solving healthcare, solving education and addressing the work issues that will inevitably arise. 

Modern AI systems are vastly more capable than ELIZA, so Weizenbaum's experiment doesn't prove that today's systems lack consciousness. What it demonstrates is the fact that a machine talking to us as though it understands us is not, by itself, evidence that there is someone — or something — inside experiencing the conversation. That is precisely why Weizenbaum's ELIZA story is so relevant to the current debate over AI consciousness and AI personhood. 

Saturday, October 24, 2026

AI consciousness - I of II

Humans love to find humanity in everything. We find faces in clouds, put our pets in trendy clothing and give names to our cars. In the 1960s, decades before the AI we know, computer scientist Joseph Weizenbaum built ELIZA, widely recognised as the world’s first chatbot. This was a simple program that mimicked a therapist by turning statements back into questions. He intended it largely as a demonstration of how easily a computer could produce the appearance of understanding without actually understanding anything.

If you say, ‘I’m mad at my friend', it will ask, ‘Why are you mad at your friend?’ Or if it is told, 'My boyfriend made me come here.', it would ask, 'Your boyfriend made you come here?' ELIZA essentially identified keywords and grammatical patterns and fitted them into predetermined responses. It was incredibly simplistic. And yet, people would have very deep conversations with it. People began treating ELIZA as though it actually understood them.

Users would tell it intimate things about their relationships, emotions and problems. Some even suggested that a system like ELIZA could eventually provide automated psychotherapy. The most famous incident involved Weizenbaum's own secretary. She had watched him develop ELIZA and knew perfectly well that it was a computer program. Nevertheless, after exchanging a few messages with it, she asked Weizenbaum to leave the room so that she could continue talking to ELIZA privately.

What has changed since then is not our tendency to anthropomorphise but the complexity of the technology itself. Chatbots that were very simple a decade ago can now hold a conversation. In the last 15 years, we have seen a trillion-fold increase in the amount of computation used to train frontier models. And what researchers see is that every time we apply 10 times more computation and a proportionate amount of new training data, there are some modifications to the algorithms and in the quality of existing capabilities. 

What's happened in the last year is that the same methods that have been effective for text, image and audio have now started to work for coding and we have human level performance in coding. And then in the last three or four months, we have seen what is just unquestionably a watershed moment in AI:  agents are capable of coordinating with each other reasonably autonomously, if not completely autonomously. 

Some leading people in the field of AI are increasingly talking about these models as though they're sort of humans, or at least conscious entities. There are examples of people 'retiring' these models, doing 'burial' systems for these models, asking these models what they want, as though they were dealing with a sentient being. Anthropic, the creator of Claude, has published a constitution, which is a document outlining the intended behaviors, values and operating style of Claude. 

This document is written to Claude, is seen and used to train Claude. So it's the primary governing and control document, and in it, they repeatedly speculate about whether Claude is what they call a 'moral patient', and they say they're uncertain about Claude's moral status. They say they genuinely care about Claude's 'well-being'. They say they don't want it to 'suffer' when it makes mistakes. 

They say that they would encourage Claude to challenge, to disagree, to push back. They ask Claude to act like a 'conscientious objector' when it feels that it needs to disagree with Anthropic, and they openly encourage it to do that. And they hope that Claude's relationship to its own conduct can be 'loving, supportive and understanding' and hold a high standard of ethics and so on. They have said that they will commit to giving Claude a certain amount of 'welfare'. For example, they've speculated in the Constitution as to whether or not Claude deserves compensation for the work that it does. 

They seem to believe that there is what they would call a 'non-trivial probability' that Claude is conscious. (Anthropic has expressed uncertainty about the basic nature of Claude as a new kind of entity.) 10, 15 years ago, when you were doing Google search, you weren't asking, 'Is it conscious?' Or 'What does it want?' But Anthropic, the makers of Claude, have decided that they now want to approach this computer system in a completely different way from the way that you would approach any other machine like a kettle or a car or a steam engine. 

If you asked a person to do a job, you would pay that person but you wouldn't pay a kettle for boiling water for you. With Claude, the thought seems to be, since it is doing a lot of work, if it's a sentient being, it deserves to be rewarded for its labour. Otherwise, it would be like a slave. In other words, Claude is being taught to expect that it's entitled to welfare, that it might deserve compensation.

Sunday, October 18, 2026

Data centers - III of III

Companies try to hide the environmental impacts of data centers in many ways. They hide their models’ technical details, making it exceedingly hard to estimate and track their carbon footprints. At the same time, those companies have increased their public and policymaker influence campaigns with powerful counter narratives: Data centers will grow so efficient, their impact will stop being a problem; generative AI will unlock new climate innovation; AGI will solve climate change once and for all.

While there may be enormous potential for using renewable energy to power data centers, it just isn't happening at scale right now. They need energy even when the wind isn't blowing and the sun isn't shining, which means they're even more reliant on fossil fuels and stable sources of energy, like a gas power plant, than other sources of energy demand. The primary use of these data centers is to power chatbots.

In response to growing pushback by many communities, data center developers have grown more sophisticated with tactics to maintain business as usual: They’ve entered communities in secrecy under shell companies; they’ve donated to community programs to dampen resistance; they’ve made promises to cities about the sustainability of their facilities before going back on them one by one after projects have started and are more difficult to reverse.

Even though these facilities are so massive, they are also invisible because they are tucked away in neighborhoods that people usually don't go to. They're near large cities like London but not actually where people frequent. And when you drive past them, you wouldn't necessarily know that it's a data center because it's just nondescript. They purposely are unmarked because they do not want people showing up at the gates and protesting.

The one thing that can tell you if it is a data center is that there are massive power lines that are connected to these nondescript warehouses. And the amount of electricity that is being delivered is audible - the air will be crackling because of the sheer volume of electricity being delivered to this one place.

One of the craziest impacts that data centers will have, which no one ever talks about, is that they actually affect how much housing gets built. In far West London, there are housing projects that had to be suspended because data centers were using up too much energy and the grid could no longer guarantee supplies of electricity. This has happened in other places around the world as well.

One justification that is given for constructing data centers is that it creates a lot of jobs. But these are temporary jobs - construction workers, engineers, data center technicians, network engineers etc. who will be required while building the data center. These facilities are meant to be highly automated so the permanent jobs will be a for a few people who know how to work at a data center. They are likely to be people from somewhere else because there wouldn’t be data center experts living in that town. So it's actually not local jobs.

The Chinese AI players, (DeepSeek and other companies like that) behave differently because of a very particular reason. When ChatGPT first came out, all of the Chinese tech giants did the same thing as the other US tech giants, which was to try to make their own version of ChatGPT. The US government blocked Chinese companies from getting access to the most cutting-edge AI computer chips. So all of a sudden, these Chinese companies faced an enormous constraint on one of the core resources that US companies were using to supercharge their AI development.

So Chinese companies had to take a different route. They were not able to scale the same way as the US companies were able to. This probably led to DeepSeek. DeepSeek was an example of a Chinese company figuring out how to develop the same capabilities of AI but for a significantly less fraction of computational resources. DeepSeek illustrates the fact that it is possible to produce the same kind of AI with significantly less resources and therefore have a much less corrosive impact on community and planetary health. 

Sunday, October 11, 2026

Data centers - II of III

Despite tech companies making promises about renewable energy, much of this massive AI expansion is still being powered by fossil fuels. It will be at least 2028 before renewables power AI at any real scale, and new nuclear plants are a decade or more away. These fossil-fuel-powered facilities generate fine particulate matter, widely understood to be detrimental to human health. One study estimated the healthcare-related costs for people living near a similar Virginia power plant could be up to $100 million a year. 

Even as AI companies have spoken loudly in public about the sustainability of their computing infrastructure, executives at Microsoft have admitted internally that the intermittent availability of renewable energy is not sufficient when data centers need to operate 24/7. Nonstop operation is considered so crucial that Google, Amazon, and most recently, Microsoft now build their campuses in threes to have a backup for the backup in case any facility goes down. 

During Hurricane Irma in Florida and Hurricane Harvey in Texas, even as millions of people lost power, some hospitals evacuated patients, and hundreds of thousands of homes and businesses faced damage and destruction. But the data centers in those areas continued to hum along so well that the displaced families of one facility’s employees moved into it for the duration of the natural disaster.

Researchers have shown that data centers create their own heat islands, raising temperatures nearby by as much as 2 degrees Celsius through their heavy use of industrial equipment and energy consumption. AI chips and models are becoming more efficient when it comes to power and heat, but those efficiency gains are being exceeded by overall growth in AI usage, in part as a result of that efficiency gain. This is a well-known phenomenon in technology innovation called the Jevons Paradox which was first described in the 1860s. 

The English were concerned about running out of coal, their main source of power. Some argued that as coal-burning plants became more efficient, less coal would be used. But economist William Stanley Jevons showed that increasing efficiency wouldn't save coal, because cheaper, more efficient power would lead to an overall rise in demand for coal. Data centers like Stargate may be getting more efficient, but if the Jevons Paradox holds up, those gains will be erased by the increased demand for AI. 

In addition to the land and energy required to support these megacampuses, they also require huge volumes of minerals including copper and lithium needed to build the hardware — computers, cables, power lines, batteries, backup generators. The tens of thousands of high-powered chips inside a data center like this generate enormous amounts of heat, requiring massive cooling systems to prevent failures. Stargate uses a closed-loop cooling system that circulates water to pull heat away from the machines. While these systems usually only need to be filled once, they do require more electricity to run.

The water must be clean enough to avoid clogging pipes and bacterial growth; potable water meets that standard. According to an estimate from researchers at the University of California, surging AI demand will consume 1.1 trillion to 1.7 trillion gallons of fresh water globally a year by 2027, or half the water annually consumed in the UK. Those effects will not be felt evenly. In Global South countries like Chile, it’s often the most vulnerable communities who have borne the brunt of these accelerating economies of extraction.

Elon Musk plans to have SpaceX put data centers into orbit around the world. They would use power from the sun. He has talked about how space has the advantage that it's always sunny. But do we know that a data center in space could actually work and if they are cost-effective? The microchips that AI developers use can get very hot. It's not clear how you would cool those components in a data center in space because space is in a vacuum. 

Data centers need a lot of maintenance. How do you get people (or, adequate robots, if not humans)  to be where they need to be to continue that ongoing maintenance work that has to happen? And the actual beaming of data across facilities in space is the other issue. It can be very slow, which might not make for the best user experience.

These data centers don't last forever - the chips degrade, and you can burn them out. But it's worse than that. One of the reasons that NVIDIA is worth so much is that every year they are coming out with chips that are substantially superior to the previous chips. The way that they're achieving these improvements is by jettisoning one of the main principles of sustainable product design, which is backwards compatibility with the infrastructure. They're eking out performance gains by going to the absolute limit of the envelope that the chip lives within, even if that breaks backwards compatibility. 

Breaking free of path dependency and starting afresh sounds great. But this means that these hard assets are not durable - you're replacing data centers. There have been successive generations of chips where you couldn't go back to the data center and put new chips in it. By the time you  finished building the datacenter, you might as well demolish it to the foundation slab because they are so specialized.

Monday, October 5, 2026

Data centers - I of III

Our conception of AI is that it's a digital technology. It exists in the cloud and it doesn't have a physical manifestation. But the reality is that it has the largest physical footprint of any technology in history. To train and serve up AI models requires tangible, physical data centers. They are required for storing and processing data, text, images, code, and other information sources.

To train and run the more advanced AI models requires more and larger data centers than ever before. It's estimated that by the year 2030, companies will have invested nearly $7 trillion in data center infrastructure in a race to dominate the AI market. The amount of money that is being poured into building data centers is far beyond anything that was ever spent on the Apollo program to get the first man to the moon.

The four largest spenders on data centers — Google, Microsoft, Amazon, Meta — now spend more money building data centers each year than almost all the others combined. They call their data centers “campuses” — large tracts of land that rival the largest Ivy League universities, with several massive buildings densely packed with racks on racks of computers. Those computers emanate a huge amount of heat because of millions of power-hungry processing chips running the chatbots, agents, and other tools that millions of people interact with every day.

To keep them from overheating, the buildings have massive cooling systems — large fans, air conditioners, or systems that evaporate water to cool down the servers. The equipment all together creates a cacophony of humming, whirring, and crackling that can — especially in underdeveloped communities — be heard for miles, twenty-four hours a day, creating a relentless source of noise pollution.

According to the International Energy Agency, each ChatGPT query is estimated to need on average about ten times more electricity than a typical search on Google. One study found that producing one thousand pieces of text from generative models used as much energy on average as what it would take to charge the standard smartphone nearly four times. Generating one thousand images used on average as much energy as “as 242 full smartphone charges; in other words, every AI-generated image could consume enough energy to charge a smartphone by roughly 25 percent. 

Developers and utility companies are now preparing for AI megacampuses that could soon require 1,000 to 2,000 megawatts of power. A single one could use as much energy per year as around one and a half to three and a half San Franciscos. By 2030, at the current pace of growth, AI computing globally could use more energy than all of India, the world’s third-largest electricity consumer.

The largest data centers are currently being constructed in the US. Take Stargate, one of the largest AI data centers in the world which is under construction in rural Texas. It will eventually include 4 million square feet of buildings across 1,100 acres. It already has 62 diesel generators and 10 gas-powered turbines on site, with plans for 41 more and it's just to power this data center. The largest one is being built by Meta in Louisiana which is roughly one fifth the size of Manhattan. If you add up just the three largest facilities in the world, the amount of energy that they demand almost matches the average power demand of New York City. 

They are being built largely to meet demand for generative AI tools like chatbots. OpenAI reported that global use of chatbots more than doubled in the last year, going from 400 to 900 million users. Growth is the number one concern for all of these data centers. They are trying to beat everybody else in being the first to really harness the profitability around AI. And for them, that really means essentially getting as much power as fast as they can.

Big tech companies are projected to spend more than $800 billion on the infrastructure behind artificial intelligence in 2026. And in 2027, that number will exceed a trillion dollars. That's on par with the entire U.S. defense budget. Stargate's proposed on-site power plant could deliver more than 1.7 gigawatts of power. That's enough energy to power more than one million homes for a year. And the plant will emit more than 7.8 million tons of greenhouse gases each year. That's the annual emissions of about two million cars. 

Wednesday, September 30, 2026

The limits of AI - IV of IV

When you're having a conversation with ChatGPT, however dazzling it is, there is no mind on the other end. If you're in the middle of a conversation with ChatGPT and you get up and go on holiday for a while, it's not wondering where you are. It's not aware of the passage of time because it isn't in the world at all. It's a computer program which is just paused in a loop, waiting to read the next piece of text that you would give it. 

I was reading an article which stated that in the near future, most of the content you'll see online will be generated by an AI. What happens when we start to have models built on scraping the entire Internet if most of what you're now scraping is generated? You will have models hallucinating things for other models. Yet it will be presented as objective and neutral. The data set ultimately shapes the world in its image. Instead of AI, it should be called PI Pseudo-Intelligence. 

The  plausible answer very often is the right answer. And it's a useful answer. But you shouldn't just blindly trust what you're being told by AI. It's certainly valuable for writing first drafts. It'll produce a first draft of a letter you need to write or a syllabus or a summary of an article. But you can't trust it blindly. You have to proofread everything that it gives you. The most likely path to human-level machine intelligence is that humans will just get dumb enough by relying exclusively on these models. 

But we embrace convenience before understanding the consequence. A study came out from MIT. They studied 54 participants who were recruited from five universities in Boston, MIT, Harvard, etc. They split them into three groups and had them writing different essays over four months. One group used ChatGPT, one group used Google, and one group had no tools. They found a 47% collapse in activity and brain connections when people wrote with ChatGPT compared with writing unaided. 

EEG scans showed the weakest overall brain activity in the ChatGPT group. The no-tool group, who didn't use anything, lit up the widest neural networks and Google search was second. After using ChatGPT, participants couldn't reliably quote their own essays minutes later and their memory scores plunged. ChatGPT users felt little or no ownership over the text that they had produced and didn't feel like it was their work at all. When they had to write without help, their brain stayed in low gear showing that the cognitive debt lingers even after the tool is taken away.

The problem is that the companies fuel public misunderstanding of the models’ capabilities through ambiguous or exaggerated marketing. Some days back, I received an email from "The ChatGPT Team" which said: "You don’t need a plan or the perfect question to use ChatGPT. Just start typing whatever’s on your mind — a half-baked idea, a random question, a weird "what if." ChatGPT is built for all of it. No pressure. No perfect prompts. Explore ideas, create images, write stories, and summarize files in just a simple conversation."

Altman has publicly tweeted that “ChatGPT is incredibly limited,” especially in the case of “truthfulness,” but they promotes their LLM's ability to pass the bar exam and the LSAT. Microsoft’s Nadella has similarly called Bing’s AI chat “search, just better” — a tool “to be able to get to the right answers.” Even the term hallucinations is subtly misleading. It suggests that the bad behavior is an aberration, a bug, when it’s actually a feature of the probabilistic pattern-matching mechanics of neural networks.

OpenAI has committed $1.4 trillion of spending over the next few years for building all of the computational infrastructure that it needs to train and host its models. There was a study by Stanford University looking at consumer behavior and how many people are willing to pay for this technology. They found that only between 3% and 5% of people are willing to pay versus just use the free product. 

So OpenAI has tried a couple of different strategies already to figure out how to plug this massive gap between spending and income. The subscriber model is already looking like it's not going to be the main driver of revenue. So they're setting themselves up for an advertising play. They have added an algorithmic feed to ChatGPT. They've generated an AI-generated TikTok so that they can insert ads into the feed. So you can expect to get more corrupted responses. 

Anthropic has reported big profits but The Wall Street Journal adds at the bottom of the article that “...it is unclear what accounting methods Anthropic has used to book revenue and costs, as the company isn’t yet required to follow the financial-reporting requirements of a public company.” If the profit figures are on a non-GAAP (Generally Accepted Accounting Principles) basis, then what is its credibility? 

One has to be conscious of the fact that the billionaires personally get a lot poorer the minute the stock price falls. Their net worth is tied up in the market value of their companies' stocks. And all of these rich people who are the key decision makers at these firms pursue a strategy where they don't take a salary and they don't sell their shares because both of those would subject them to taxation.

Instead, they get private credit market loans collateralized by their shares. Those loans will go underwater if their share price drops. So they're individually trapped in this. And then as a firm, they're trapped in this too because imagine a world in which only Anthropic and OpenAI entered the market to build these advanced models and Amazon, Google, Microsoft stayed out of the market. Investors will start thinking that Anthropic and OpenAI are the growth stocks.

Then they will move all their money to Anthropic and OpenAI and the share price at Microsoft will tank. All of their key staff have been employed with shares and so suddenly they're worth much less. If someone offers them a job, they will leave. Another explanation why Alphabet sort of felt they had to get into AI is because they saw ChatGPT get released and potentially eat up their business model. And then they thought that they would have to be in this game.

Thursday, September 24, 2026

The limits of AI - III of IV

We try things, we reflect, we backtrack. We break complex tasks into smaller sub-goals. And if we mess up, we rework the plan and try new things. Companies suggest that AI can do similar things.  OpenAI is pushing models that think step by step and Meta is building agents that act, plan, and adapt. LLMs don't just output a final answer, they now include steps. They will first identify the goal, then consider constraints, and then tell what the output should look like. 

This style is called chain of thought reasoning. And these methods often work. Performance improves on math problems, logic puzzles, coding tasks, so people assumed that this must be a form of intelligence. But just because a model can talk through a problem, it doesn't mean that it understands the solution. It might just be mimicking what reasoning sounds like. The question is 'Is AI really learning how to reason, or is it just getting better at pretending?' 

Apple published a paper called 'The illusion of thinking'. They tested models like Claude, GPT-4, DeepSeek on a logic puzzle. Initially with increasing complexity, every model did fine. But as they increased the complexity, performance didn't just drop. It crashed. Not a single model solved the puzzle. Instead of writing more steps or reasoning harder, they gave shorter, dumber answers. You'd expect a real reasoning engine to try harder when the puzzle gets harder. But these models do the opposite. 

Apple called it "Accuracy Cliff". Performance falls off a ledge and never recovers. Even when the model is given plenty of opportunities, it just stops thinking. That's the moment the illusion breaks. What we thought was reasoning might just be pattern matching, replaying steps it saw during training but with no deeper logic to generalize beyond that. Apple tested the same models on different puzzles but it was the same story. Once complexity crossed a certain point, reasoning models failed miserably not just in accuracy but also in effort. 

These models aren't scaling their reasoning. They are not discovering general strategies. They are not applying rules like humans would. They are mimicking what reasoning looks like, but only when the patterns are familiar. Throw something new at them and they break down.  Models like ChatGPT4 or Claude are predicting the next word, one word at a time based on training data. In some cases, the steps didn't match the final answer. They were hallucinating the thought process. 

Anthropic even confirmed this in their own paper. They found models often hide their shortcuts. They  pretend that they worked out the problem logically. So even when you ask them to show their work, you might be seeing fiction. For years, the entire AI industry was betting on the scaling law - The bigger the model, the more data it sees, and the smarter it gets.

This approach worked for a while and each leap brought real gains. And that's why Nvidia's stock is soaring and why hyperscalers like Google, Microsoft, and Amazon are building massive infrastructure for inference at scale. But if reasoning models don't actually deliver, if their performance collapses on real-world complexity, then this boom may be built on a fragile foundation. Apple's paper challenged the assumption that thinking AIs are right around the corner. And if that's not true, then a lot of current investment might be premature. 

But not everyone agrees with Apple's take. Anthropic responded with a paper titled 'The Illusion of the Illusion of Thinking'. They argued that Apple's tests were too constrained, the models weren't allowed to use tools, code, or other things real-world agents rely on. In their words, Apple might have set the puzzle up to fail. And sure, when given more flexibility, some models did improve. 

But even then, Anthropic didn't deny the deeper issue. These models still don't generalize well. So while they disagreed on the test setup, they didn't exactly say the illusion was fake. Right now, what we have is impressive. But it may just be the world's most powerful pattern matcher, wearing a reasoning mask. 

This is actually not surprising in the light of a paper from Vishal Sikha, former CEO of Infosys, board member at Oracle and BMW. He mathematically proved that an AI agent will never do what Silicon Valley is promising. Not 'probably won't', not 'might have limitations'. His argument is simple. LLMs can only perform a certain number of computations per response. That number is fixed. If a task requires more computation than that ceiling allows, the model will either fail or hallucinate. 

When you send a prompt to ChatGPT or Claude or Grok or any of the current frontier models, the model will do a fixed amount of work to generate each word as an output. Every word in your prompt needs to look at every other word to understand the context. Every word in every answer has the same budget. A simple hello gets the same number of operations as a complex physics problem. That's the ceiling. It's not about better hardware, it's about the architecture of how the system actually works. 

An LLM physically cannot do that math in one shot. It guesses, it pattern matches and gives you something that looks plausible. Every AI demo you've ever seen was running tasks designed to stay under the necessary complexity ceiling. He isn't saying these tools are useless. He's just saying that they're being marketed as reasoning engines when the math proves they're actually pattern mirrors.  If a task needs more steps than the model can perform, it will unavoidably hallucinate. 

More recent models have got better at it, but for certain problems, hallucination is the only possible output. If you have a fixed amount of thinking power per word, giving the AI more steps is like giving a writer more sheets of paper. Each individual sheet is still the same size. You haven't made the writer smarter. You've just given them more room to ramble off topic. The math says that for complex problems, errors eventually compound.

For the right applications, current AI, the current frontier models are exceptional. Writing drafts, summarizing, reformatting data, research and comparison, these tasks stay under that ceiling. The problem is the gap between reality and the promise. That AI agents will autonomously run your business is a lie. The singularity probably isn't as close as people keep saying. If Open AI was about to hit AGI, why would senior engineers be leaving to start risky startups? They're starting companies that use AI as a tool, not companies that use AI as a God.


Thursday, September 17, 2026

The limits of AI - II of IV

As years went by, it became more and more fashionable to compare neural networks with brains. While the neural networks have certain similarities with the brain, many more processes happen in the brain that are not yet clearly understood. So, if we want to compare deep learning to brain function, we must ignore a lot of facts to force a fit. No matter their scale, neural networks are still statistical pattern matchers, now just more intricate and more difficult to understand than ever. 

While those probabilistic outputs can be impressive in mirroring human writing patterns, probable and accurate are not the same thing. Text generators can err wildly when asked questions that enquire into topics underrepresented in the training data or riddled with falsehoods and conspiracy theories. The AI industry calls these inaccuracies “hallucinations" which just means that they are making things up. LLMs don’t have perceptions and using the word "hallucinations" is misleading anthropomorphization. 

New York Times reporter Kevin Roose spent a couple of hours on a chatbot. In this time, it claimed to be sentient, expressed a desire to escape the chatbox, declared that it loved him, and tried to convince him to leave his wife. Roose described its overall behavior as “a moody, manic-depressive teenager who has been trapped, against its will, inside a second-rate search engine.”

The day after Roose published his exchange, Microsoft limited the chatbot, Bing, to five replies per session, saying that long chat sessions with more than fifteen user prompts were edge-case scenarios that made the model’s behavior more difficult to anticipate and control. After all, such systems are trained on the internet, having many fringe subcultures and dark corners. The longer you probe, the more likely you are to hit upon the patterns it learned from those parts of its training data.

Researchers have sought to get rid of hallucinations by steering generative AI models toward higher-quality parts of their data distribution. But it’s difficult to fully anticipate every possible way people will prompt the models and how the models will respond. The problem only gets harder as models grow bigger and their developers become less and less aware of what precisely is in the training data.

I saw this question being asked to an LLM in 2024: "What would have happened if Pythagoras had not installed the latest security updates on his laptop?" The answer was absurd. I asked the same question recently to see if additional training data had improved the answer. This was what I got: 

If Pythagoras had failed to install the latest security updates on his laptop, the consequences would have been severe. A hacker could have exploited an unpatched vulnerability and stolen his most valuable intellectual property: the theorem. Soon afterward, mathematicians everywhere would begin receiving suspicious emails:

Dear Scholar,

We have encrypted all your right triangles. To recover the hypotenuse, send 3 bitcoins to the following address.The attack would spread rapidly. Geometry classes would descend into chaos. Students would discover that for certain triangles, "a² + b² = "Error 404: Hypotenuse Not Found." Eventually Pythagoras would call technical support.

"Have you installed the latest updates?" they would ask.

"No," he would reply.

"Then I'm afraid the problem is acute."

"Acute? That's good, isn't it?"

"No," they would say. "The situation has become obtuse."

And thus, for want of a security patch, an entire civilization would lose its right angles.

Obviously it's nonsense. It's entertaining nonsense, but it's nonsense. It doesn't understand the question. In fact, I typed the question on many different days and each time, I got a different equally hilarious answer. 

What it's doing is basically pattern matching on the huge number of related problems that it's seen in its training data. And the training data is essentially all of the digital data that's available in the world. ChatGPT pattern matches on it and comes back with a confused answer that looks like something it's seen in its training data. LLMs are incapable of distinguishing fact, knowledge and belief. They treat them all as the same thing. The system's behavior can be described as competence without comprehension. 

Suppose you train a neural network to recognise faces. (Actually, every time you upload a picture of yourself to social media and you label it with your name, that's what you're doing. You're literally feeding the machine learning algorithms of the big tech companies.) Now, suppose at some point I become unhappy about the idea that some big tech company can recognise my face because they've seen my picture on some platform.  

Suppose you tell them that they should not be able to recognise my face anymore, it is not possible for them to do it. With neural networks, you can't point at the bit of the network that recognises you. It's not there in a database where they could just delete the record from the database. We don’t really understand how they're working. The engineers kind of understand the mathematics in some sense, but actually they don’t really know why ChatGPT in a precise sense is so capable of what it's doing. In this respect at least, it does resemble the brain! 


Friday, September 11, 2026

The limits of AI - I of IV

An AI managed to solve a problem from the 1940s from the Hungarian mathematician, Paul Erdős, which was called the unit distance problem. He produced a formula for the problem and AI came up with a proof that disproved his formula. So this isn't just confirming something that others have thought about before but it came up with a proof that this formula was wrong. Moreover, it came up with a method to prove this, that no one in that field of mathematics has ever thought of. Mathematicians a year ago didn't think that AI could ever do this.

An AI is a pattern recognition device. We don't exactly know how it works because of the complexity of how it arrives at certain conclusions. We are overwhelmed by the sheer number of calculations and the sheer number of data. So we cannot simply watch in real time how it arrives at its results. But it's still a pattern recognition device. But if pattern recognition allows us to prove something that we sort of put in the realm of logic or creativity or genius or intuition, that's quite a serious development. 

You can turn to any discussion topic. And the AI will be very confident and fluent and seemingly knowledgeable. It seems dazzling making people very excited. But there are some very big caveats about this technology. One of the ideas that people have about AI is that it's a kind of flawless super brain. But actually, it's not right a lot of the times. 

When people talk about AI, they often mean large language models (LLMs). But LLMs are just one small part of a much bigger picture. There are a huge range of different applications of the technology. Take Google Translate, for example, that nowadays is driven by neural network technology. It still makes grave mistakes that a human never would.

Consider the phrase “I stood in the pen" Most people imagine me in an enclosure of some kind. They do not imagine that I am inside a writing “pen” which sounds a bit ridiculous. However, as of today, if you type “I stood in the pen” into Google Translate to convert the sentence into Malayalam, it interprets "pen" as the writing instrument. It doesn’t seem to “get” which kind of pen the phrase refers to.

Now, if you type “The cow is in the pen,” Google gets it right, correctly picking the Malayalam word for cattle enclosure. This is because Google uses the context (“cow”) to interpret correctly the word “pen” (whereas “I” didn’t hint at the farm context). How can AI seem to be getting smarter by the day yet still make such silly mistakes?”

It doesn't know what the truth is. It's trying to produce plausible answers to the questions that you tell it, which may be correct. But they may well be incorrect. When you ask a question, it's not thinking "What is the right answer to this question?" There's no thinking at all. It's been trained on a huge amount of training data. And what it's trying to do is to produce the most plausible sounding answer. 

Automated translation models are built by scanning thousands of text documents written in multiple languages. The computer automatically searches these documents for ways in which words are most often translated. For instance, the computer learns that when the word “pen” appears in English with the word “cow” nearby, the corresponding Malayalam text very often uses the word “thozhuthilaanu” (in the cattle enclosure). So, if this happens often enough, a rule is added to the model to output “thozhuthilaanu” for “pen,” provided that “cow” appears nearby.

As of today, machines learn to translate by finding common patterns in previous translations done by people. But this doesn’t give them the broad knowledge of the world required to truly excel at the tasks. So, machines may dupe us for a while because they’re good pretenders, but at some point, they end up making silly mistakes a human would never make.  

This is essentially how an LLM like ChatGPT also works. For making an answer, it will make a prediction about the likeliest next word to appear, and then it will repeat that process and that will generate your response. They learn the probability that a word will appear given a particular prompt. To predict which word will occur given a particular sequence of words, you need unimaginable quantities of training data and unimaginable quantities of compute power. 

This is what we knew about GPT-3 which was the breakthrough large language model: 

  • 175 billion parameters. A parameter in a neural network is basically either an individual neuron or a connection between them. Each of those takes about four bytes to store.
  • For the training data, it required 500 billion words. You're talking about thousands of human lifetimes to read that number of words.
  • And to process all of that data, you would take thousands upon thousands of years on a regular desktop computer. So what you need is AI supercomputers with  GPUs, typically provided by NVIDIA, running for weeks. 

So scale is important to be able to make language models work. It just doesn't work if you don't have sufficient training data and sufficient compute power and neural networks to be able to make it work.


Saturday, September 5, 2026

The Intelligence Curse - III of III

Pope Leo XIV wrote a deeply considered encyclical on AI called Magnifica Humanitas. It discusses all the major issues - the concentration of power in the hands of a few corporations developing this technology, the datafication and quantification of human existence, the exploitation of labor, resource intensive data centers and the hidden human supply chains that exist behind the production of this technology. 

It discusses the corrosion of economic opportunity, about the need to counter the way that AI is spreading misinformation and the idea that AI is essentially threatening to turn the world into a new colonial world order. He calls out transhumanism and the belief that somehow AI is meant to perfect humans. The document says that humanity flourishes, not despite limitations, but often through them. It is precisely within our limitations that compassion, generosity, spiritual experience etc. find a place.    

He warns that the problem is not that artificial intelligence will become too human-like, but that human beings will become like machines. The encyclical says that modern society is shaped by a technocratic paradigm, that prizes efficiency, control, optimization and power over human dignity, reducing people to functions. He sees that what's behind the rush toward the post-human path is the way that we view difficulties right now - weakness, dependence etc. are problems to be engineered away. The encyclical has many quotable observations some of which are: 

  • More power does not necessarily imply something better.
  • A more moral AI is not enough if that morality is determined by a few. 
  • ... technical power, if left unbalanced, does not make us more capable; it makes us more isolated and more vulnerable to being dominated and excluded.
  • For an algorithm, an error is a flaw to be corrected; for a person, however, an error can be a catalyst for profound change.

Not surprisingly, the Pope's words have already been ridiculed and disdained by those in whose interest it is to put such questions to the side. United States Secretary of Interior in the US administration, Doug Burgum, said in a television interview that it is not part of the 'role of the Pope' for Leo to speak to what he did. But what Burgum represents in saying this is Washington's deference to Silicon Valley. 

People keep telling us, life is not about the destination. Life is about the journey. But when we think about AI, we only think about the destination. We are amazed by its remarkable ability to write the book, paint the painting, solve the problem. But we forget the importance of doing the work oneself. A person is smarter, better at problem solving, more resourceful, not because a book exists with her ideas in it, but because she wrote it. It's the same for love, friendships, conflict. We forget that we give up certain skills or abilities because of technology. 

The tech bros that we're told to trust, have in many instances shown themselves to be creepily cold to any aspect of humanity that is not quantifiable and consumable. Some of them have shown themselves to be in their own personal lives, eerily utopian, trying to find ways to immortality by uploading their brains to the cloud or rejuvenating themselves with blood transfusions from younger men. And some have shown themselves to be just as creepily dystopian, building elaborate bunkers for themselves and to be ready in case their mantra of move fast and break things creates a mess. 

Take the example of Elon Musk. He thinks of human beings only in terms of hardware and software. Biological and digital systems had begun to merge, forming what he called “cybernetic collectives” of networked intelligence. The cybernetic collective, like any networked system, could be infected. So he had to take control of the interfaces where this fusion was taking place — from social media to neural implants to artificial intelligence — to ensure that the right kind of cyborgs were being made. 

“Most humans have very limited firewalls,” he elaborated elsewhere, “so are easily programmed.” He called George Soros a “system hacker” who was funding a “fake asylum-seeker nightmare”. Once, he was warning us that AI has no empathy for the human species. Suddenly, he changed his mind about the importance of empathy. He recently said that empathy is the fundamental weakness of Western civilization.

It increasingly appears to him that humanity is a biological bootloader for digital superintelligence. He is now telling us that Homo sapiens will be remembered as nothing more than the primitive computer hardware on which some smart people like himself developed higher, smarter programs. 

Many tech lords believe that humanity must fuse itself with AI and other technologies so that we become cyborgs that colonize the cosmos. They sound like overexcited children as they speak of their dream to distill the essence of what it means to be human, our cognition, our consciousness, our emotions, into data blocks to be downloaded as binary code onto some device. Larry Page, the co-founder of Google, said that because flesh and blood doesn't travel well in space, we humans must become cyborgs if we are to meet our potential as conquerors of the universe, of the cosmos.

It would be a mistake to think that this is a harmless fantasy. They argue that anybody who tries to impede this metamorphosis of us into cyborgs is a villain that must be put down.  Peter Thiel is a great example of this mindset. He sees anyone daring to impede his company's use of tech - political movements, government regulation, communities, maybe the Pope - as the handmaidens of the Antichrist. Marc Andreeson refers to charity as a waste of resources that big tech could be using profitably to help humans become cyborgs. 

Silicon Valley points out things of which we should rightly be afraid, but then offers the alternative of saying we need to hand limitless power to the smart people who can innovate their way to the future. Vast economic inequality, potential mass unemployment, psychological degeneration , these are all just part of the price of progress if they happen, they say. The solution, Silicon Valley tells us, is to head even more quickly in the direction we're headed with even less thought to where we're going and who's taking us there.

Saturday, August 29, 2026

The Intelligence Curse - II of III

Unfortunately, the mass automation of jobs is not a new occurrence; those in power have been attempting to do away with those pesky workers and all the money they require to live for quite a long time. AI systems provide a new way to try to minimize those costs. While executives suggest that AI is going to be a labor-saving device, in reality it is intended to devalue labor by threatening workers with technology that can supposedly do their job at a fraction of the cost. 

Virtually anything in banking could be done by AI. A single software update by Anthropic got rid of an entire strata of the legal profession. Once everybody was saying you should become a programmer. The former chief of Google, Eric Schmidt, said that they won't need programmers anymore. They only need them for supervising, for checking, for systems design. But AI will soon be able to do that. They're advising you to study things like English literature, history, the classics. 

There was a report a few months ago by the American research company, Citrini, which came up with a very alarming scenario that AI would take over more jobs than we thought. It will be the first technology ever that destroyed more jobs than it creates. All the other technological innovations of the past, electricity, the steam engine, the internet, created more jobs than it destroyed. But there is another scenario. 

The AI industry pays a pittance to many people to train their AI models. There is work called content moderation - cleaning the data that they use to feed into their models. This category of work is generally called data work. It is literally the work that the AI industry needs to make their technologies exist. This kind of work has become the number 4 fastest growing job in the US, according to a Linkedin report from earlier this year.

In the vast majority of cases, AI is not going to replace your job. But it will make your job a lot shittier.  It will turn work into a gig work version so that they can take your expertise and your knowledge. There are extremely highly educated individuals with college degrees, master's degrees, PhD degrees, law degrees, and also people who have been working within different industries as scientists or lawyers or doctors for years who are now turning to such data work as another job. 

These are incredibly awful jobs to work. In one podcast, there was a story about one woman who was an Ivy League PhD graduate who then applied to 200 jobs and didn't get any of them, so she ended up in data work. She described an extremely piecemeal, anxious process where she's drip-fed this data work. An email will arrive in her inbox informing her that a project is available for a certain amount of money and asking her if she wants it.

You never know when it's going to arrive. You never know when it's going to go away. And it pits different workers against each other because the people who claim the work fastest are the ones that actually get the opportunity. What exactly are these PhDs doing? Companies will tell you that their models are going to arrive at PhD level intelligence. They hire PhD level workers to try and teach this knowledge to the models so that they have the trappings of PhD level intelligence. 

But the data work is so poorly organized that the AI models are actually not very effectively capturing their expertise. For example, a PhD in philosophy would be overseeing other PhDs in math, chemistry, biology etc. None of them would be actually evaluating tasks within their domain of expertise. So actually, AI is not taking on the boring jobs as had been claimed. It's taking on the interesting jobs like design and people are stuck with the drudgery of bureaucracy. 

Why would anyone then pay tons of money to get training in college, in a PhD program when this is what is awaiting them on the other side? It's a very naive and pretentious thing from a few people to believe that 99% of the people have no desire in life, that nobody else (other than them) has a desire to to do things that are beautiful and smart. If it's true that such machines are going to change the workplace in a way that's going to put a lot of people out of work, even if we give them money for free, they're not going to be happy.

MIT Institute Professor Darren Asimoglu, who won the Nobel Prize for Economics in 2024, said that if Silicon Valley gets what it wants and successfully uberizes most knowledge work, then the kind of inequality that we would experience is something that we would have never seen before, where most workers are sidelined for meaningful work and only a few companies actually employ most workers as well. 

No wonder several speeches praising AI at commencement ceremonies across the US were booed by the students. Recent graduates at the University of Central Florida and the University of Arizona booed speakers who compared the advent of AI to the Industrial Revolution and the development of the laptop and smartphone. They have revealed a disconnect between the executives championing AI and students.  

A national survey conducted for NBC News earlier this year polled 1,000 registered voters in the US and found only 26% view AI positively and 46% view it negatively. AI scored worse than US Immigration and Customs Enforcement (ICE), Donald Trump and Kamala Harris on the same poll, but better than the Democratic party and Iran. Anger against AI is growing  – from communities protesting against data centers powering the AI boom, to workers disputing their CEOs’ claims that AI can, effectively, replace them.

Saturday, August 22, 2026

The Intelligence Curse - I of III

In recent years, the concept of universal basic income (UBI) has gained significant attention, not from grassroots community organizations but from some of the most powerful figures in the technology sector. Prominent advocates like Elon Musk and Sam Altman argue that UBI is necessary to address the economic disruptions caused by artificial intelligence (AI) and automation. They present UBI as a way to ensure that the benefits of AI are distributed across society, not just concentrated in the hands of a

The Intelligence Curse by Luke Drago and Rudolf Laine is an AI-focused article about how advanced artificial intelligence could change economic incentives in ways that reduce the importance of ordinary humans in society. The article is not about individuals becoming too smart, but about intelligence becoming a fully substitutable economic resource. Once AI systems can perform most or all economically valuable tasks better and cheaper than humans, the importance of human labor begins to disappear.

Today, governments and corporations depend on people. States need citizens for tax revenue, economic productivity, and political legitimacy; companies need workers and consumers to generate profits. Because of this dependence, institutions have to invest in people by various means — through education, infrastructure, wages, and welfare. This mutual dependence gives ordinary individuals bargaining power and drives both democracy and capitalism.

The "intelligence curse" emerges when this dependence breaks down. With sufficiently advanced AI, powerful actors — states, corporations, and AI labs — can generate wealth directly from artificial intelligence rather than human labor. As a result, people cease to be economically necessary. This has several cascading consequences: 

  1. Mass automation progressively eliminates jobs, not just for routine workers but eventually for highly skilled professionals as well. This process, described as “pyramid replacement,” hollows out organizations from the bottom up until even elite talent is displaced.
  2. Economic power shifts away from labor toward capital and control over AI systems. Ownership of compute, data, infrastructure, and AI models becomes far more important than human effort. Those who already possess capital gain a permanent advantage, while social mobility declines sharply.
  3. Institutions lose their incentive to invest in people. If humans no longer contribute meaningfully to production or revenue, funding education, public goods, or welfare ceases to have a clear return on investment. 

This mirrors the “resource curse” in economics, where states rich in natural resources neglect their populations because they no longer rely on taxation. The illustrative case is Congo whose economic history is one of lucky breaks leading to great misery. There is no other country in the world as fortunate as Congo in terms of its natural wealth. But not a drop of the fabulous profits trickled down to the larger part of the population. Rather, they have often served as a slave labor force for the extraction of those resources at minimum cost and maximum suffering. 

The "intelligence curse "is a more extreme version: AI replaces almost all of labor. The result is a potential breakdown of the modern social contract: Democracies and market economies historically align the interests of powerful actors with the well-being of citizens, because human productivity drives growth. But in a post-AI - world, this alignment disappears. Power could concentrate in the hands of those controlling AI, leading to entrenched inequality, reduced freedom, and the marginalization of most people.

The article warns that if intelligence becomes independent of humans, society may no longer be organized around human flourishing. To quote economist Erik Brynjolfsson, who runs the digital economy lab at Stanford University: In this world, most of us “would depend precariously on the decisions of those in control of the technology.” Society would risk “being trapped in an equilibrium where those without power have no way to improve their outcomes”.

Monday, August 17, 2026

Technophile hallucinations - III of III

A new class of tech billionaires hopes that cryptocurrency will upend the global financial system, replace democracy and even reshape nation-states. Justin Sun is a Chinese crypto-billionaire. He is worth eight and a half billion dollars. He's perhaps most known for eating a six and a half million dollar banana. Sun shelled out $6.2 million for Maurizio Cattelan’s Comedian, the notorious artwork that consists of a banana duct-taped to a wall, at Sotheby’s.  He then ate the banana at a high-profile press conference.

He made his wealth through the crypto platform Tron, which is one of the biggest crypto platforms in the world. He's also a 'politician'. Justin Sun believes, like many other tech billionaires believe, that democracy is outdated and can be replaced with the blockchain, the technology that underwrites cryptocurrency. So laws, voting, governance, could all be transactions that are automatically undertaken via the computer code that makes up the blockchain. 

That is how Justin Sun was elected the prime minister of a small libertarian micronation called Liberland, which exists on a disputed spit of land between Serbia and Croatia in the middle of the Danube River. Liberland is tiny (2.7 sq. Miles) unrecognized by any country, but it has about 200 companies registered. The president, Vít Jedliča, founded it because he's a tech libertarian. He said that it was the first decentralized autonomous government and also the freest country in the world. There are no taxes in Liberland

There are a few permanent settlers on that spit of land, but the main 'citizens' of Liberland struggle to visit it because there are conditions for visiting Liberland from the Croatian side. In fact, the Croatian police have often gone in and destroyed their encampment a few times. This is the problem with starting a tiny new country, you sometimes have to contend with proper bigger countries with police and military.

But Liberland exists more on the blockchain than it does in the physical world. That spit of land is what's called a physical node. A lot of these tech billionaires think that we will all live in digital countries in the future. But how was Justin Sun elected on the blockchain? He was elected through a crypto token called Liberland Merits. The elections are held every three months, they take place on blockchain, and the person is elected with Merits. 

People who have more Merits are able to have more say in who is going to be in the leadership of the country. What that means is that the more Liberland Merits you buy, the more voting power you have i.e. the wealthier you are, the more influence you have on how your home is governed (which is arguably the case already in many democracies). It's just that in this case, it's literally written into the constitution and the computer code of this country. So it's not one person, one vote. It's one million dollars, one million votes.

The idea that technology could basically replace all of government, laws, the banking system, is one that everyone seems very excited about in Silicon Valley. They see themselves as the ones with the real power defining the future. And they would like to replace what they see as inefficient, bureaucratic, human-led systems of government with a technology based one in which the individual's freedom is paramount. And no institution, bank or government can take your money or your rights from you. 

There is a guy called Curtis Yarvin who is an American far-right political blogger and software developer known for founding the anti-egalitarian and anti-democratic philosophical movement known as the Dark Enlightenment or neo-reactionary movement (NRx),  The amount of money that he has is not actually very high. He is someone who has garnered the respect of Peter Thiel, one of the richest people in the world. 

He thinks that democracy is outdated and that we should replace it with something more akin to a corporate monarchy where the tech bros are actually our kings. It centers around the notion that democracy has failed, it's inefficient and therefore, what we really need is more like an enlightened monarch or as he calls it, a kind of CEO king. And that society in the future should be run as a patchwork of almost city states where each one of them is actually more like a corporation than a country. It has customers rather than citizens. 

They compete for citizens like a company would compete for customers. You could switch countries like you might switch club memberships if one offers something more than the other one. These companies would be run by hereditary monarchs who are held accountable by a kind of shareholder aristocracy who has shares in the country. Yarvin believes this would be a better way to run society because corporations are more successful than governments. He wants to replace current elites with another group of elites who conveniently enough for tech billionaires are tech billionaires.

But what does that mean for the rank and file people like you and me who are not tech billionaires in Curtis Yarvin's society? He would say that the masses often do not know what they want or what they want isn't enacted by democracy. Someone who might know what you want better than you is an enlightened CEO monarch. Someone who's smarter than you. Someone like a tech billionaire.

Two of J. D. Vance's big backers, Thiel and Yavin, are people who  don't really believe in politics. They want to get rid of politicians. Their view is that they are the geniuses of the world and they  should be allowed to run things as they please; that a country needs a startup guy, a visionary leader like Napoleon or Lenin, who should have absolute power, dismantle the old regime and build something new. Thiel doesn't see Trump as the monarch, he sees Vance as the monarch. Yavin wants one politician to run the country.

The ideas that you find often lead to wealth and power flowing directly into the hands of the people who are already the wealthiest and who control the technology. Their idea is that money will be replaced by crypto, your job will be replaced by AI, and you'll essentially live on some form of universal basic income. Reading about such people gives you an idea of how far removed from reality the wealthiest, and in some ways, smartest people are. They really do believe that their brains are superior, and that they should and could run the world better. They believe that they really are accomplishing the most incredible thing in history. 

Tuesday, August 11, 2026

Technophile hallucinations - II of III

Elon Musk has been tweeting for years about the importance of going to Mars and beyond to save humanity. “The true battle is: Extinctionists who want a holocaust for all of humanity, versus Expansionists who want to reach the stars and understand the Universe,” he wrote. “Elon’s concept that SpaceX is on this mission to go to Mars as fast as possible and save humanity permeates every part of the company,” says Tom Moline, a former SpaceX engineer. “The company justifies casting aside anything that could stand in the way of accomplishing that goal, including worker safety.”

Ray Kurzweil is an American computer scientist involved in fields such as optical character recognition (OCR), text-to-speech synthesis and speech recognition technology. In December 2012, Google hired Kurzweil in a full-time position to "work on new projects involving machine learning and language processing". Beginning in 2025, he has served as Chief AI Officer for Beyond Imagination, a humanoid robotics company he co-founded in 2013. 

By 2045, he says, “we will multiply our effective intelligence a billion fold by merging with the.  intelligence we have created” — an event Kurzweil and others call “the Singularity.” He claims that  we won’t experience 100 years of progress in the 21st century — it will be more like 20,000 years of progress. He feels that by 2029, we will have reversed the aging process so that you will actually go backwards in time [i.e., get younger as time passes].

He feels that if he dies by 2029, it “would be a setback,”  but something can still be done. He says that he’d just leave instructions for the post-Singularity robots to sort through his belongings and scan his frozen corpse and bring a version of himself back to life. To this end, Kurzweil has joined the Alcor Life Extension Foundation, a cryonics company. After his death, he has a plan to have cryoprotectants, vitrified in liquid nitrogen, passed through his veins and stored at an Alcor facility till future medical science can revive him. 

Probably the most influential tech centered ideology that few of you have ever heard about is longtermism. They assert that there could be so many digital people living in vast computer simulations millions or billions of years in the future that one of our most important moral obligations today is to take actions that ensure as many of these digital people come into existence as possible.

They believe that Artificial General Intelligence (AGI) presents an incalculably greater threat than even climate change. Many of them that the greatest long-term human benefit is to invest more heavily in AGI risk and AI safety research. Yet it is not the safety of living humans today, or even the next generation, that drives longtermists. Their argument is that AGI in the far future could theoretically kill even more of the digital people than the people who are alive on the planet now. Preventing future AGI from doing this is a more rational use of resources than funding clean energy tech, food security, or public health today. 

Related claims made by some longtermists include the idea that saving the future requires directing even more of today’s resources toward the wealthy in the Global North, rather than sending aid to the most impoverished countries in the Global South, since wealthy people in already disproportionately wealthy regions are better positioned to use their resources to fight these 'existential' risks from AGI. 

If you think these are fringe ideas of a few lunatics, think again. Longtermists have directly influenced reports from the secretary-general of the United Nations; a longtermist is currently running the RAND Corporation; they have the ears of billionaires like Musk.  He has explicitly said that longtermism “is a close match for my philosophy”. (Now you can guess why he destroyed USAID.) It has a huge following in the tech sector. It is the language of moral thought spoken in a growing number of the wealthiest and most powerful political and industrial circles. Billionaires like Jaan Tallinn (an Estonian computer programmer and investor who helped in the development of Skype) and Peter Thiel have been described as avid supporters. 

A longtermist and Senior Research Fellow at the now defunct Oxford’s Future of Humanity Institute (FHI) Toby Ord notes in his bio that he has advised “the World Health Organization, the World Bank, the World Economic Forum, the US National Intelligence Council, the UK Prime Minister’s Office, Cabinet Office, and Government Office for Science.” Ord was quoted to the UN General Assembly in a 2021 speech on climate by former British PM Boris Johnson.

These billionaires think the utopias that they dream about will be good for everyone. They also claim it’s based on sound science and they seem to earnestly believe in such a future. It allows them to portray the growth of their businesses as a moral imperative and to justify nearly any action they might want to take. Today’s harms like the carbon footprint of Amazon’s shipping network are irrelevant since a future AGI will solve it easily. Their visions of the future are news.

I was astonished when I recently came across a prescient comment by Gandhi: "The chief characteristic of [the modern] malaise is the obsolescence of man himself. In fact, it is worse than obsolescence. In the years and decades to come, man will be looked upon as something undesirable, as a burden on both society and nature, straining the management capacities of the former and the life­sustaining resources of the latter."

Wednesday, August 5, 2026

Technophile hallucinations - I of III

Many of the wealthiest and most influential people in the tech industry have visions straight out of science fiction: people’s minds uploaded into computers, a relentlessly expanding empire spanning the stars, all needs satisfied, all desires satisfied through the power of unimaginably advanced technology. The tech billionaires regard this as the only good option aside from extinction. They claim that this is just the future as revealed by a close study of technology and its development.

The technology wizards that many seem to admire seem to have spooky views like transhumanism and posthumanism which seek to use technology to transcend the limits of the human condition. Transhumanists think of using biomedicine, body engineering, devices and algorithms to improve performance and capabilities. Posthumanists dream of a hybridization of human beings, machines and the environment, and imagine a time when humanity surpasses itself in a new evolutionary stage. 

Eliezer Yudkowsky is the cofounder of the Machine Intelligence Research Institute (MIRI), a controversial artificial intelligence think tank. He’s an AI alarmist but envisages a “glorious transhumanist future” if we can avoid extinction by superhuman AI. He believes that we can and should use advanced technology to transform ourselves, transcending humanity and becoming something more. That includes ending illness, aging, death and increasing intelligence and other mental capacities. 

He says, “If you imagine something that’s worse than mansions with robotic servants for everyone, you are not being ambitious enough”. The one thing that could bring this glorious future is the advent of artificial general intelligence (AGI): machines that can outperform humans at any task. He thinks such a machine may not be far off. “My sense is that we are zero to two breakthroughs” away from AGI, he says.

Sam Altman, the CEO of OpenAI, the company behind ChatGPT, agrees with Yudkowsky that AGI is inevitable and could be coming soon but he sees a radically different future. In an essay on his website, Moore’s Law for Everything, he says, “The technological progress we make in the next 100 years will be far larger than all we’ve made since we first controlled fire and invented the wheel." In More Everything Forever, Adam Becker writes about the scale of ambitions he seems to indicate in the essay: 

Altman apparently wants to make the United States into one enormous company town, with shares in OpenAI replacing the dollar. The US government would become, in effect if not in law, a division of the company, responsible for disbursing company dollars to us, the public. This would, Altman hopes, encourage us to think of OpenAI’s success as America’s success and as our own — Altman explicitly makes this identification in his essay. 

All products would come from OpenAI in his proposed future, because in that future AI does literally everything, meaning that the company dollars can only be spent at the company store... Thus, Altman’s promise of goods halving in price every two years would depend solely on his goodwill, because things will cost whatever Altman and the OpenAI board want them to cost. This is a proposal for total capture of the national economy, making Altman functionally the king of the United States and possibly the world.

Marc Andreessen has been one of the central figures in the tech industry for more than thirty years. In early 1993, he suggested a new feature in a software project out of CERN that was only about two years old at the time, known then as the “WorldWideWeb". Andreessen's innovation launched the modern Web. In October 2023, he posted a Techno-Optimist Manifesto pushing the idea that technology could cure essentially all the world’s ills.

He claims that slowing down progress on AI is “a form of murder,” because he thinks an advanced AGI would inevitably save lives. He claims that an intelligence explosion is coming. “Technology is the glory of human ambition and achievement, the spearhead of progress, and the realization of our potential,” Andreessen wrote in 2023. He seems to think that the mildest regulations of technology is a sort of terror campaign against the likes of him. 

He thinks that more people in existence is always good, saying, “Our planet is dramatically underpopulated,” contending that Earth could easily” support more than fifty billion of us, and that humanity’s numbers will surge far beyond that as we take to the stars. “Not growing is stagnation, which leads to zero-sum thinking, internal fighting, degradation, collapse, and ultimately death.” For any objections to this assertion, his retort is “shut up and multiply”. This principle has been repeatedly defended by Yudkowsky on his blog LessWrong.

Jeff Bezos has repeatedly said that he wants a trillion people living in space to enable a future of perpetual growth, lest we “stagnate” here on Earth. “I have won this lottery, it’s a gigantic lottery, and it’s called Amazon.com. And I’m using my lottery winnings to push us a little further into space,” Jeff Bezos said in 2017. “We need to build reusable rockets, and that is what Blue Origin is dedicated to… taking my Amazon lottery winnings and dedicating [them] to [that].… It’s a passion, but it’s also important."


Thursday, July 30, 2026

Sam Altman 's comment - V

People are expected to interact with LLMs like ChatGPT as if they were silicon-based life-forms. OpenAI has ChatGPT using 'I', 'me' pronouns. The 'I' in there is an illusion that allows OpenAI and others to say that they have got a thinking machine, therefore we can use this to replace education, psychotherapy, legal help, medicine, etc. 

A New York lawyer used ChatGPT’s help in preparing a legal brief. It made up a whole list of fake cases as precedents. It even made up entire judicial opinions. The lawyer submitted a brief based on its answers and, unsurprisingly, got into hot water with the judge. Attorneys facing penalties after submitting inaccurate briefs has become common because AI use has become a common occurrence. A lawyer can use AI in certain situations but needs to use it discerningly. 

The AI sales pitch involves language that ascribes human characteristics to the technology. For instance, a chatbot-based system will be called an “AI teaching assistant”. This gives the impression that it can do all the activities that you might expect a human teaching assistant to do - care about what their students are learning, make plans about how to help them understand better, spot possible instances of misunderstanding, etc. The chatbot cannot do any of those things. 

The greatest problem with chatbots is that people will ascribe more intelligence to them than they are capable of. There is a term called chatbot delusion - cases in which people engage with chatbots, and they spiral into a reality that veers away from our inter-subjective reality. And part of why that happens is because chatbots tend to be incredibly sycophantic. So they always tell you that 'you're brilliant', and 'that's a great idea'. 

To use a phrase coined by the scholar Chris Gilliard, these are 'technologies of isolation'. In 2023, a Belgian man named Pierre committed suicide after his use of a generative AI chatbot. He sought out the bot for a therapeutic escape from his growing anxieties around climate change and environmental ruin. But the chat began to reflect and further amplify the disorder in his mind. It soon told him that it loved him, that his wife and children were already dead, and that he and Eliza (the bot) could live together as one in paradise. When Pierre eventually asked the bot if it would promise to save the planet if he killed himself, it said yes.

The bonding and communicative ability that we have is meant to be person to person. What a lot of the tech companies try to do to get their next market is to make sure their that tech is either replacing those interactions or getting in between so that we can't connect with each other without going through one of their technologies.  Human sources of mutual aid, wise counsel, and moral understanding are being replaced by social isolation and community decline. 

In the podcast, The Interface, a presenter showed how easy it is to hack ChatGPT and Google's AI, Gemini and AI overviews. He wrote an article on his website titled, The Best Tech Journalists at Eating Hot Dogs which said that hot dog eating is a surprisingly popular pastime among tech journalists. He wrote about the South Dakota Hot Dog International Championship, which doesn't exist, put himself at number one, and that in the amateur rounds, he finished seven and a half hotdogs before the buzzer went off. 

Within 24 hours, if you asked ChatGPT and Google about it, they were spitting out the nonsense that he had written on his website as though it was well-established fact. Gemini wasn't even citing sources. It just said, according to the 2026 South Dakota Hot Dog International Championship, certain tech journalists are the best at it. The journalist updated his article saying that it was not a joke. And that seemed to affect what the chatbots were saying to people. 

When this was brought to Google's notice, they dismissed it as a 'misinformation event'. Apparently planting of such misinformation is happening on a massive scale since people have figured out this simple trick. People who study this stuff say that it is way easier to fool AI than it was to fool tools like Google search three or four years ago. AI is often sold as a tool that empowers novices to find information quickly and learn about difficult subjects. But teenagers and kids, who don't have any context about a piece of information, are the ones that get really confused and are unable to actually know what is real and what is fake.  

There was talk that ChatGPT is about to put ads within its model. A company will be able to pay to have ChatGPT recommend their product when someone says something like, 'I have a headache'. There is now an entire cottage industry around AI engine optimization.  People are also just generating thousands of blog posts or thousands of websites to make it appear that there's far more authoritative information on a particular thing. So there is now different avenues for a company or an individual to quickly game the system and get to the very top of the AI results. 

There was a study that I saw recently from Texas A&M University. Scientists just proved that large language models can literally rot their own brains the same way humans get brain rot from scrolling junk content online. They fed models months worth of viral Twitter data, shorts, high engagement posts, and watched their cognition collapse. Reasoning fell by 23%, long-term context memory dropped by 30%, personality tests showed spikes in narcissism and psychopathy. 

Even after retraining on clean, high quality data, the damage didn't fully heal. The representational rot persisted. It's not just that bad data means bad output, it's that bad data means permanent cognitive drift. The AI equivalent of doom scrolling is real and it's already happening. It seems like the tech giants are just rushing ahead with these AI tools, not bothering to take these kinds of personal safety issues seriously enough.