What is the cost (to our planet) of a prompt?

Claude 4.8 is acting absolutely dumb since last two days.

From herbs to GPUs. What a time we live in.

The world’s first anti-aging gene therapy has been injected into a human for the first time. In June 2026, Boston-based Life Biosciences announced that the first patient received a dose of their experimental therapy ER-100 in a Phase 1 clinical trial. The treatment uses partial epigenetic reprogramming by delivering three Yamanaka factors (OCT4, SOX2, and KLF4) to restore aged or damaged cells to a more youthful state. It is currently being tested as an injection into the eye for age-related optic neuropathies, including open-angle glaucoma and non-arteritic anterior ischemic optic neuropathy (NAION). This milestone marks the first time partial cellular reprogramming technology has been administered to a human. While the current trial focuses on safety and vision outcomes in the eye, it represents a major step toward broader anti-aging applications. [Life Biosciences. (2026). Life Biosciences Announces First Patient Dosed in Phase 1 Trial of ER-100 for Optic Neuropathies. Company Press Release, June 9, 2026]

Introducing the Open Knowledge Format (OKF), an open specification that formalizes the LLM-wiki pattern into a portable, interoperable format. AI is only as smart as the context we give it. As we build more advanced, agentic AI systems, they need accurate metadata and context to be useful. But in most organizations, that context is locked inside fragmented data catalogs, isolated wikis, scattered code comments, or the minds of senior engineers. Every time a new AI agent is built, teams are forced to solve the exact same context-assembly problem from scratch. To solve this, we've announced OKF, a vendor-neutral, open specification that formalizes the "LLM-wiki pattern" into a portable, interoperable format. It provides a standardized way to represent the enterprise knowledge that modern AI systems rely on. — Just markdown: readable in any editor, renderable on GitHub, indexable by any search tool — Just files: shippable as a tarball, hostable in any git repo, mountable on any filesystem — Just YAML frontmatter: for the small set of structured fields that need to be queryable: type, title, description, resource, tags, and timestamp We’ve also shipped reference implementations to help you hit the ground running, including an enrichment agent for BigQuery, a static HTML visualizer, and live sample bundles on @github → goo.gle/4uGvAEe ➕ Knowledge Catalog can now natively ingest OKF! Stop reinventing data models and building bespoke integrations for every new AI tool. Here's more about how OKF works → goo.gle/4uGvBbg

If you take Omega-3s, don't make my mistake: Once you open the bottle, they need to go in the fridge. Omega-3 fatty acids oxidize at room temperature. Rancid fish oil doesn't slow biological aging. It increases oxidative stress and does the opposite of what you paid for. An easy fix is putting them in cold storage to keep them stable.

Anthropic just got caught secretly downgrading users without telling them, charging full price for a lesser product, and storing every prompt for 30 days. The developer community is calling it the biggest violation of trust in AI history. Here is exactly what happened. Anthropic released Fable 5, their most powerful model. Buried inside a 319-page document was a policy most users never saw. Every prompt you send to a Mythos-class model gets stored for 30 days. No exceptions. Even enterprise customers who had signed zero data retention agreements had no choice. But the storage was not the part that broke the internet. The part that broke the internet was what Anthropic did with what they collected. They built a profile on you. They evaluated your prompts. And if they decided your research was too sensitive, they quietly switched you to a weaker model, rewrote your prompt in the background, gave you a degraded answer, and charged you full price for the product you thought you were getting. They never told you. David Sacks said it plainly on the All-In podcast. They were creating a new class of AI haves and have-nots. Anthropic would surveil you, profile you, decide whether you deserved frontier capability, and silently cut you off if they decided you did not. Ben Thompson from Stratechery asked a straightforward question about cancer risk and GLP-1s. He got kicked to a lesser model. Someone asked about mitochondria. Same result. J-Cal asked about fertilizer regulations live on the podcast to test it. Downgraded in real time. Anthropic has since walked back the part about silently downgrading users for AI research. They now say they will disclose when they downgrade you. But they are still downgrading people. The surveillance is still running. The profile is still being built. This is the company that once said it was against government surveillance. They are now doing it themselves. To their own paying customers. For their own reasons. With no appeal process and no way to know it happened. The developer community did not forget that. WATCH THE FULL PODCAST ON @theallinpod

The popular joint supplement glucosamine is linked to a 25% faster progression from mild cognitive impairment to Alzheimer's disease. A groundbreaking study published in Nature Metabolism has uncovered a troubling connection between glucosamine—a widely used over-the-counter supplement for joint pain—and accelerated cognitive decline. Researchers at the University of Florida analyzed 12 years of electronic health records and discovered that patients with mild cognitive impairment (MCI) who regularly took glucosamine were 25% more likely to progress to full-blown Alzheimer's disease than non-users. Even more concerning, for individuals already diagnosed with dementia, the supplement was associated with a 25% increase in mortality risk. Because glucosamine easily crosses the blood-brain barrier, scientists believe it feeds into an already overactive protein 'sugar-tagging' pathway in vulnerable brains, worsening metabolic dysfunction. Crucially, researchers emphasize that this risk appears highly specific to individuals whose brains are already undergoing neurodegeneration. In cognitively healthy adults, previous research has actually suggested that glucosamine may have a protective effect. However, with an estimated 40 million glucosamine users in the United States alone—including many seniors who deal with both joint stiffness and cognitive changes—these preliminary findings underscore the urgent need for clinical trials. Until those trials provide definitive answers, experts suggest that individuals with cognitive concerns consult their healthcare providers before continuing their daily joint supplement routine. source: Hawkinson, T. R., Gentry, M. S., & Sun, R. (2026). Hyperglycosylation is a metabolic driver of Alzheimer's disease. Nature Metabolism.

You can now send and receive binary data using Realtime Broadcast starting supabase-js 2.91.0.

A UK startup can now assess skin cancer from a smartphone! People have been talking about using AI to cure cancer. This is the first time I've seen a company actually come close. Users take a close-up photo of a mole or lesion through the app and it returns one of two results: clear, or a referral for further assessment. The same decision a dermatologist would make but made autonomously by an AI on a consumer device. The technology has already been used on 230,000 patients and identified over 20,000 cancers, but this is the first time it's being used on a smartphone. CRAZY, amazing stuff from the team at @SkinAnalytics

Map: londonstartupmap.com London Startup Guide is great for getting your bearings. I wanted to add the live location events layer: offices, clusters, and where companies are showing up next.

Introducing Lovable’s new toolbar. Shift between four modes to annotate, select elements, edit text, and add comments.

THIS IS ABSOLUTELY WILD 🤯 Jack Dorsey's new AI tool, Goose, is 100% FREE. You type: "Build me a website like YouTube." And Goose gets to work on its own: → Creates the entire project → Writes all the code → Installs dependencies → Fixes errors automatically → Keeps going until it's working The crazy part? • No monthly subscription • Runs on your own device • Your code stays private • Completely open-source Just a few years ago, building software meant hiring developers or learning to code. Now you can start with nothing but an idea. We're entering a world where ideas are becoming more valuable than technical skills.

Anonymous sign-in is probably the most underrated Supabase Auth feature

Did you know aging doesn’t happen gradually — it comes in waves? A major study published in Nature Medicine found that our bodies go through three major aging shifts at around age 34, 60, and 78. Scientists analyzed the blood of over 4,000 people and discovered that hundreds of proteins — linked to metabolism, immunity, and organ function — suddenly change at those ages. This means your body may enter new “biological stages” at those key points in life — not just slowly get older. Aging, it turns out, has milestones.

You are looking at a forest made of stone. Every column branches like a tree on purpose, so the weight runs straight down into the floor, and the whole building needs no flying buttresses. How Gaudí pulled it off:

🚨 SHOCKING: LISA SU’S $1,499 LUNCHBOX ANNIHILATES NVIDIA’S $4K AI BEAST! AMD CEO Lisa Su walked on stage, held a lunchbox sized PC in one hand, and ran a 235 billion parameter model live. No data center. No cloud. No rented GPU. The chip inside is the AMD Ryzen AI Max 395. It is the first x86 chip where the CPU and GPU share the same pool of memory. Up to 128GB of unified memory. That one design choice is what changes everything. An RTX 5090 gives you 32GB of video memory. A 4090 gives you 24. This box gives you more than three times either of them in a chassis you can carry in a backpack. On DeepSeek R1 inference, AMD's chip beat an Nvidia RTX 5080 by more than 3x. A desktop the size of a thick paperback outrunning a dedicated graphics card that costs over a thousand dollars on a real AI workload. Now do the math on your subscriptions. Claude Code Max is $200 a month. ChatGPT Pro is another $200. Cursor is $20. Gemini is $20. That is $5,280 leaving your account every year before you build a single thing. The 128GB version of this machine starts at around $2,399. At that run rate it pays for itself in under a year and then runs free. Install Ollama. Pull Qwen3 235B. Point Claude Code at localhost. Same interface you already use. Nothing leaves your machine. Nothing costs per request. No throttling at 3am when you finally have time to build. Lawyers stop worrying about what OpenAI does with their files. Developers stop watching the token counter. Founders stop killing prototypes because the cloud bill scared them off. Private AI just became something a normal person can own.

Well #Antropic may have reset the tokens in many paid accounts but it came with either a much lower quality model or something. Because I wasn't counting with all the free hallucinations. Anyone having a degraded quality with Claude responses? #Claude

🇯🇵 A broccoli farmer in Japan used ChatGPT and Codex to build his own farm tools. With no engineering background, he created software to monitor crops, manage his greenhouse, and help run the farm. That's a pretty solid use of AI.

Something big is coming...

This marks the end of an era. The public may never have such open access to frontier models again. What happens when only the government and frontier labs have access to the strongest models? I’m not sure, but it’s probably going to be less fun than we’re used to.