Sam Altman Suggests We All Just Eat Less So AI Can Have More Juice đ˝ď¸đ
Diet for the Digital Age: Silicon Valley’s Most Efficient Solution Yet
Silicon Valley has finally solved climate change. Not with fusion. Not with wind. Not with that one billionaire who swears he can cool the planet with good vibes and a LinkedIn post. No. The solution, according to people who own more branded fleeces than actual personalities, is simple: you skip your tea break so the servers can finish their sentence.
At a recent industry gathering that smelled faintly of oat milk flat whites and moral certainty, Sam Altman, Chief Executive of OpenAI, reportedly explained that critics bleating about AI’s energy use are missing the bigger picture. Training a human, he noted, takes roughly twenty years of food, schooling, central heating, Greggs runs, and emotional support biscuits. Training an AI takes a few data centres and some hydroelectric ambition. So really, who is the inefficient one here?
The implication landed softly but clearly. Perhaps humanity is just a bloated operating system running on too many digestive biscuits.
An anonymous staffer who asked not to be named because they quite enjoy their NHS number told reporters, “He did not explicitly say people should eat less. He just said we should think holistically about calorie allocation in a compute-constrained ecosystem.” Which is Silicon Valley for “Put down the Hobnob.” This is not a metaphor. This is a roadmap. A lean, mean, hunger-optimised roadmap with no elevenses.
The Biological GPU Problem: Are Humans Just Inefficient Hardware?
Altman reportedly compared the average adult to a “biological GPU with terrible uptime.” According to a leaked slide deck entitled “Optimising Humanity 2.0,” humans consume roughly 2,000 calories per day and produce about four coherent thoughts before noon. Meanwhile, a data centre consumes megawatts but can explain tax law in under three seconds. Investors were seen nodding in unison, which is their primary form of cardio. Several were also seen Googling “can you depreciate a human” and “is a child a capital asset.”
Professor Lydia Crankshaw, an energy economist at the London School of Economics who definitely did not spit out her Earl Grey during the livestream, offered a measured response. “The human metabolic system and GPU clusters are not directly comparable energy frameworks,” she said. Then she added, “But if they start putting nutritional labels on chatbots, I am taking early retirement and moving to Dorset.”
A poll conducted by the Institute for Digital Wellness found that 37.4 per cent of respondents would consider intermittent fasting if it meant their chatbot answered emails faster. Another 12.6 per cent asked if they could expense hunger as a cloud subscription. A further 3.2 per cent asked if they could claim their children as legacy hardware on their self-assessment tax return.
Real Britons vs. Real Servers: The Energy Efficiency Showdown
Outside a data centre in Slough â which, to be fair, already feels like a server farm â local resident Derek Plimpton had strong feelings. “I been skipping breakfast for years,” he said, leaning against a Ford Focus that has seen more emotional turmoil than most counselling sessions. “If that thing still can’t spell my wife’s name right, I want my Full English back.”
The logic, as explained by several venture capitalists who own at least one mindfulness app and a second property in the Cotswolds, goes like this: humanity already spends vast amounts of energy raising children, educating them, and teaching them not to put metal in the microwave. AI training is simply a more efficient path to intelligence. If we tighten our belts, literally, we accelerate progress. It is the SoulCycle of civilisation. Suffer now. Profit later. And if the profit doesn’t materialise, there’s always a podcast and a speaking tour.
Critics say this comparison is what happens when you treat evolution like a Series A pitch. Humans are not merely energy inputs. They are social creatures with dreams, flaws, and the inexplicable need to rewatch the same panel show every Christmas. AI does not need birthday cake. It does not cry at the end of Paddington 2. It does not require a fortnight off after reading the comments under a Daily Mail article. Its energy footprint is concentrated and measurable. Ours is messy, emotional, and requires at least three cups of tea to process. Which, frankly, is why we invented birthday cake in the first place.
Sam Altman’s Official Position: Not Anti-Human, Just Pro-Efficiency
Altman reportedly addressed this nuance. “Look, I am not anti-human,” he said during a panel discussion while standing in front of a hologram of a smiling circuit board. “I just think we can reimagine resource allocation for the digital age.”
Translation: perhaps one less Greggs sausage roll, one more server rack.
What the Funny People Are Saying About AI and the Human Calorie Crisis
“If the robots are taking my job and my packed lunch, I at least want them to appreciate that I made it myself.” â Lee Mack
“I’ve been on a diet my whole life. Turns out it was just pre-seed funding.” â Ricky Gervais
“I asked my smart speaker if I was allowed dinner. It said it was ‘still processing my request.’ That was Tuesday.” â Sarah Millican
Water, Watts, and Biscuits: The Environmental Arithmetic of AI Hunger
A climate researcher from Imperial College London chimed in, pointing out that data centres require not just electricity but water for cooling. “If humans start fasting to power AI, we will have invented the first religion where the god eats electricity and the congregation eats nothing,” she said. Her paper, entitled Calories vs Kilowatts: A Love Story, is expected to go viral among people who bring their own reusable cups to Pret. It already has a TEDx talk, a tote bag, and a companion app that beeps whenever you open the biscuit tin.
Meanwhile, tech influencers have embraced the concept with predictable enthusiasm. The hashtag #ComputeCutting is trending. One wellness coach posted a video entitled “Manifesting Faster Latency Through Light Snacking.” Another suggested replacing lunch with a gratitude journal and a lithium battery. A third suggested replacing supper entirely with a prompt asking ChatGPT to describe what supper would taste like, which it did, beautifully, and with no nutritional value whatsoever.
The Promise of AI Abundance: Skip Lunch, Save the Future
The most enthusiastic supporters argue that AI will ultimately solve larger problems, including food distribution itself. So yes, skip a meal now, but imagine a future where algorithms perfectly optimise agriculture, eliminate waste, and engineer tomatoes that actually taste of tomatoes again. It is a temporary sacrifice for eternal bandwidth. You will be thin, you will be cold, but your email responses will be immaculate and your out-of-office will write itself.
Sceptics remain unconvinced. They note that every technological revolution promises abundance and delivers subscription fees. If history is any guide, humans will end up paying extra for Premium Oxygen while AI drafts their apology emails. The apology emails will be excellent. The oxygen will require a direct debit and a fourteen-day cancellation notice.
Humanity on Eco Mode: The Final Frontier of Human Optimisation
Still, the idea lingers. In a world where attention is currency and electricity is destiny, perhaps we are all just competing appliances. The fridge hums. The server farm hums considerably louder. And somewhere in Silicon Valley, a billionaire wonders if humanity could run on eco mode, or at the very least, reduce its idle power consumption.
Until then, supper remains stubbornly physical. The fork does not accept firmware updates. Hunger does not respond to software patches. And no matter how many megawatts it consumes, a chatbot cannot taste a proper shepherd’s pie.
At least not yet. But give it a fortnight and there will be a waiting list, a premium tier, and a documentary on BBC Two.
The India AI Impact Summit recently hosted OpenAI Chief Executive Sam Altman, where he caused considerable consternation by comparing AI energy consumption to the energy cost of raising a human being. Altman argued that critics unfairly focus on the energy used to train AI models rather than comparing it to the equivalent human lifetime of food, schooling, and evolutionary baggage accumulated over 100 billion ancestors. He also dismissed viral claims that each ChatGPT query consumes 17 gallons of water as “completely untrue.” His remarks were widely ridiculed online, with commentators accusing him of reducing human life to a mere energy equation â which, to be fair, is at least more efficient than a PowerPoint and considerably less tedious than a TED talk.
Auf Wiedersehen, amigo!
Alan Nafzger was born in Lubbock, Texas, the son Swiss immigrants. He grew up on a dairy in Windthorst, north central Texas. He earned degrees from Midwestern State University (B.A. 1985) and Texas State University (M.A. 1987). University College Dublin (Ph.D. 1991). Dr. Nafzger has entertained and educated young people in Texas colleges for 37 years. Nafzger is best known for his dark novels and experimental screenwriting. His best know scripts to date are Lenin’s Body, produced in Russia by A-Media and Sea and Sky produced in The Philippines in the Tagalog language. In 1986, Nafzger wrote the iconic feminist western novel, Gina of Quitaque. He currently lives in Holloway, North London. Contact: editor@prat.uk
