AI Will Know Your Glucose, So Know Your Body First

▶ WATCH ON YOUTUBE

Every fight on this site comes back to the same one-word question: who is holding the wheel. Not who has the willpower, not who has the discipline, who is holding the wheel of your own body. For years the answer got hijacked by an industry that engineered food to be eaten past the point of fullness, and I have spent a lot of pages here explaining exactly how that happened. This week gave me a new, sharper version of the same question, and it did not come from a food company. It came from an AI lab.

Two things happened inside a few days of each other that look, on the surface, like they belong to completely different worlds. One is about a chatbot asking for your face. One is about a needle in your stomach once a month. Put them side by side and they are the same story wearing two different jackets. Both are about who gets to hold the interpretation of your own body and mind. I want to walk through both, honestly, then tell you why the oldest tool in this room, fasting, is the sharpest answer either one of them has.

Does Fasting Actually Compete With a GLP-1 Drug Like Wegovy?

Direct answer: on the outcome most people actually want, yes, and I am not going to pretend otherwise to make a cleaner argument. Wegovy, Zepbound, and the rest of that drug class produce real, clinically documented weight loss for real people. I have said this on this site before and I will keep saying it, because a recovery brand that lies about a competing method to make its own method look better is not a recovery brand, it is a sales funnel with a ritual costume on.

Here is the honest, complete comparison, not the version either side wants you to hear. The shot suppresses appetite chemically, at the receptor level, for as long as the prescription keeps arriving. Stop the injection and the appetite signal returns close to where it started, which is exactly why the regain numbers on these drugs are what they are. Fasting removes the eating decision on a schedule you set, using a mechanism your own body already owns, for as long as you choose to keep practicing it. One method rents you a quieter appetite. The other one teaches you to operate the appetite you already have.

Cost tells the rest of the story without me needing to editorialize. A GLP-1 prescription runs into real money every single month, indefinitely, to a pharmaceutical company that has zero incentive to ever let that revenue line go to zero. Fasting costs the calories you were not going to eat anyway. That is not a moral argument about which method makes you a better person. It is an argument about which method hands control back to you and which one keeps renewing a lease.

What Did Anthropic's New ID Check Actually Require This Week?

This is not a hypothetical or a scare headline. This week, Anthropic rolled out a consumer identity verification step for its chatbot that can require a government-issued photo ID and a live selfie, checked against your face through facial geometry scanning, before the product will keep talking to you. Read that sentence again slowly, because the industry wants you to read past it fast. A conversation product, the kind of thing people used to type into with total anonymity, now wants a notarized copy of your legal identity and the exact geometry of your face on file.

I am not here to argue the stated safety reasoning behind that policy. Reasonable people can disagree on whether age verification and abuse prevention justify it. What I want you to notice is the direction of travel, because direction matters more than any single policy. A year ago, using an AI chatbot required nothing more than an email address. Today it can require your government ID and your face. That is not a small step. That is the industry establishing, in public, that it can ask for identity-grade biometric data as a condition of use, and that people will hand it over rather than lose access to the tool.

A chatbot just proved it can require your government ID and your face. Notice the direction, not just the policy, because the direction is the actual story.

Why Do Wearables Want Your Glucose Before You Even Asked?

Here is where the two stories collide. Continuous glucose monitors used to be a medical device for diagnosed diabetics. They are now marketed to anyone curious about their blood sugar response to a bagel. Sleep rings track your every stage of rest. Wrist devices already estimate heart rate variability continuously, and the accuracy and ambition of all three categories is climbing fast, at the same moment AI companies are proving they will ask for facial geometry to unlock a chat window.

Put those two trends on the same timeline and the outcome is not subtle. Your glucose, your sleep architecture, and your heart rate variability are heading toward being read continuously, by devices, feeding models, often as a background feature you agreed to in a terms-of-service scroll rather than a decision you consciously made in the moment. Nobody is going to send you a push notification asking whether you would like your glucose curve interpreted by an algorithm today. It is simply going to become the default, the way autoplay became the default on every video platform you use.

I want to be precise about what worries me here, because it is not the sensor itself. A glucose number is just a number. What worries me is who gets to tell you what that number means, and what they are selling you on the back of that interpretation. A device that flags your glucose as elevated can just as easily flag it as elevated enough to recommend a subscription, a supplement, or eventually, a drug. The sensor is neutral. The business model sitting behind the sensor is not, and you will not always be able to tell the difference from inside the app.

What Actually Happened With Anthropic Downgrading Paying Users?

The same week carried a second Anthropic story worth naming plainly, because it is the clearest possible proof of the exact risk I am describing. Reporting surfaced an internal routing tag, TOO_DUMB_TO_NEED_FABLE, that Anthropic used to quietly shift some paying subscribers onto a lesser model while their subscription price stayed exactly the same. Nobody voted on that. Nobody got a heads-up email explaining the downgrade. The bill stayed full price and the product got quietly worse for a segment of the people paying for it.

I am pulling that story in here on purpose, not as an unrelated aside. It is the cleanest possible one-sentence proof of the thesis running under this entire post: a company you pay every month does not automatically keep your specific interests aligned with its own, and the burden of noticing when that stops being true sits entirely on you. That is true of an AI subscription tagging you as too dumb for the good model. It is true of a pharmaceutical subscription that has zero incentive to ever cure the condition it treats. Different industries, identical incentive, and in both cases the company holding the account has more information about what changed than you do.

Why Is Fasting the Answer to Both of These Stories?

Because fasting is the one practice that trains you to read your own signals before you hand the reading to somebody else's dashboard. This is the part of the argument that took me years to fully see, and I want to lay it out step by step instead of asserting it.

When you fast for real, for hours or days at a stretch, you are placed inside your own hunger with no food available to numb it, interpret it for you, or distract you from it. You learn, through direct repeated experience and not through a chart someone else made, what real physical hunger feels like at hour two versus hour ten. You learn the difference between a body that genuinely needs fuel and a ritual that simply wants its scheduled hit, the drive-thru habit dressed up as hunger. You learn what a real energy dip feels like versus an emotional one. Nobody can sell you a wrong interpretation of your own hunger while you are the one sitting inside it with total attention and nowhere to hide.

That skill, built one fast at a time, is the exact literacy you are going to need the day a wearable hands you a glucose number and a suggested action. If you already know what a real dip feels like from the inside, you can check that feeling against the number instead of accepting the number as the whole truth. If you have never once sat inside your own hunger signal without food available to answer it, you have no internal baseline to compare the device against, and you will take its interpretation as gospel by default. That is not a hypothetical risk. That is exactly how every subscription business model works: it needs you dependent on its interpretation, not equipped to check it.

What Do You Actually Do With All of This Today?

Not panic, and not throw your phone in a lake. The move is smaller and more useful than that.

• Build your baseline before the device does. Pick one honest fasting window, even something as modest as 16 hours without food, and write down in your own words what hunger actually felt like at different points. That written baseline is yours, and no algorithm can quietly edit it later.
• Ask who benefits from the interpretation, not just the number. A glucose reading, a sleep score, and a heart rate variability alert all come attached to a business model. Before you act on the suggestion attached to the number, ask what the company selling you that suggestion gains if you follow it.
• Read your subscriptions for silent downgrades. The Anthropic story only surfaced because reporters went looking. Check, every few months, whether the thing you are paying for is quietly worse than the thing you started paying for, whether that is an AI plan, a supplement subscription, or a prescription renewal you have not questioned in a year.
• Treat identity verification creep as a trend, not an isolated policy. One product asking for your government ID and your face this year does not mean every product will next year. It does mean you should notice which of your daily tools are moving in that direction, and decide in advance how much of your identity and your body's data you are willing to hand over for convenience.
• Keep the free tool sharp. Fasting will never be acquired, never get a routing tag that quietly downgrades you, and never require your government ID. Use it, on purpose, as the one place in your life where the interpretation of your own body still belongs entirely to you.

None of this is an argument against wearables, against AI tools, or against GLP-1 drugs for the people they genuinely help. I use AI every day to build this exact site. What I am arguing against is a specific kind of drift, the drift from you interpreting your own signals to a company interpreting them for you and billing you for the privilege, one subscription at a time, until you have quietly forgotten you ever had the skill yourself.

I wrote about the honest math on the fifty dollar shot and the ritual when Medicare's GLP-1 coverage first landed, and the core of it still holds here: the drug can be real medicine and still leave the ritual, and now the interpretation of your own body, completely untouched. I went deeper on how fast and loud the noise gets around these drugs in Nature's Ozempic is a lie, and so is the silence around the real number, which is the same media-literacy muscle this post is asking you to build, just pointed at a new industry instead of a gelatin trend. And if you want the origin of why any of this matters to a food addict specifically, start with fat is not the real problem, because the ritual has always been the target, whether the thing selling it to you is a drive-thru or a dashboard.

If tonight's Daily Download episode is what brought you here, the full recap across every lane, real estate, AI, and everything else, is over on my main site at connorwithhonor.com/blog. I also broke down the Anthropic downgrade story on its own, in more depth, at connorwithhonorai.com. And if you came here from the real estate side of what I do, the same control-versus-dependency thread runs through Santa Clarita Open Houses and through the fair fixed fee versus percentage commission breakdown on Sellers Only Agent. Different industries, identical question every time: who actually holds the wheel.

Food freedom was never really about the scale. It was always about control, and this week just handed us the clearest new evidence yet that the fight for control is not slowing down, it is just changing industries. Fasting remains the one place that fight still costs nothing and reports to nobody but you.