Discrediting work you don't understand isn't a criticism. It's a confession.
There's a type of comment I've started to recognize before I even finish reading it.
It comes in the first few hours of a video doing well. It says something like "she just uses AI" or "ChatGPT told her this" or my personal favorite, "this is so irresponsible for the environment." It's always confident. It always arrives from someone who has spent more time on TikTok today than I have spent using Claude this entire week. And it always reveals the same thing, not that what I'm doing is wrong, but that the person typing it has no idea what I'm actually doing.
That's the thing I want to sit with. Not that they disagree with me. Not that they're hostile. But that they genuinely, completely, do not understand what this process is, and they're commenting on it anyway like they do.
So let me explain it. All of it. Because once you understand what's actually happening here, the criticism doesn't just fall apart. It exposes itself.
What you don't see
Every night, I spend hours doing yoga and sitting with what I'm noticing in my own body.
Not what someone told me to feel. Not what a prompt suggested I try. What is actually, physically, undeniably changing, in my fascia, in my face, in the way pressure moves through my system when I work in a specific place. I notice something shift. I go back to it. I document what I felt. I feel it again. I stay with it until I understand it well enough to know it's real and repeatable, before I ever bring it anywhere else.
That's where this started. In December. With a sensation I didn't have language for. Something releasing in my body and I didn't know the anatomical name for what I was feeling. I just knew something real was happening, and I kept returning to it because the results were visible, repeatable, and unlike anything I had seen anyone else doing or describing.
The feeling came first. Always. The feeling comes first.
What you see on camera, the technique, the name, the explanation, is the end of a process that starts in my body in the dark, hours before anyone else is involved.
What Claude actually does, and why the "she just uses AI" take embarrasses itself
Here's the gap that has always existed in this space and that nobody has closed.
Anyone can show how they stretch. Anyone can post a before and after. What almost nobody can do is accurately translate what they are feeling in their own body into language that helps someone else find it in theirs.
That is the actual problem I'm solving. And it's bigger than people realize.
There has been no mass revolution in stretching, in fascia work, in body-based practice, not because the techniques don't work, but because people don't have the vocabulary to bridge their internal experience and someone else's. You feel something shift and you don't have a word for it. You notice a change and you can't explain the mechanism. So you either stay silent or you guess, and neither one helps anyone replicate what you found.
I use Claude to close that gap. To take what I felt in my body, cross-reference it against anatomy and medical literature, and give it a name that is accurate, real, and transferable. Not because I need a fancy label. Because language is how felt experience becomes something other people can find in themselves.
Claude is a research tool. It is not an idea generator. It does not tell me what to feel or what to try. It identifies, names, and cross-references what I already discovered, against published anatomy, against medical literature, against a knowledge base that would take years to accumulate manually.
The distinction is not subtle. Using Claude to research your discoveries is exactly like using PubMed to verify a hypothesis you formed in a lab. The experiment came first. The literature comes after. That is what research is.
The people saying "she just uses AI" are describing the last five percent of the process and presenting it as the whole thing. They're looking at the citation and saying the experiment never happened. But I wasn't watching my body from the outside and asking Claude to explain it. I was feeling my body from the inside, and using Claude to translate those feelings into real science.
That's not the same thing. That's not even close to the same thing.
And here's what I genuinely want to know: show me someone else doing that. Show me one other person in this space taking their internal physical experience, verifying it frame by frame, cross-referencing it against medical literature, and delivering the result with enough precision that someone on the other side of a screen can feel it in their own body.
The criticism isn't that I'm doing something wrong. It's that they don't understand what I'm doing. And those are very different problems, only one of which is mine to fix.
On Claude's actual accuracy, because this matters
Since we're here, let me be specific. Because vague defenses don't help anyone and the people making this criticism have clearly never looked at what Claude actually is.
Claude is a large language model trained on an enormous body of text including medical literature, anatomy textbooks, peer-reviewed research, and clinical documentation. It has access to PubMed's 35+ million biomedical articles. It is not a random chatbot producing plausible-sounding nonsense.
In formal academic testing on anatomy questions, USMLE-style multiple choice questions covering seven anatomical topics, Claude scored 76.7% accuracy. The previous generation of AI scored 44.4% on the same questions. Random guessing scores 19.4%. These are not comparable performance levels.
In anatomical recognition tasks using medical imaging datasets, Claude 3.5 Sonnet demonstrated the highest anatomical recognition accuracy of any model tested at 85%, and was rated the most consistent model overall, 83% consistency in anatomy identification and 92% consistency in fracture detection across repeated tests.
That last part matters more than people realize. Consistency means when I bring the same observation back from a different angle on a different day, I get structurally grounded answers that hold up, not random variation. It means the anatomy I'm naming is stable and cross-referenced, not invented.
Is Claude a licensed anatomist? No. Is it an extraordinarily powerful tool for cross-referencing a physical observation against the deepest body of medical knowledge ever assembled? Absolutely yes.
There is no more accurate way to do what I'm doing. Frame-by-frame footage of my own body, cross-referenced against a model trained on more anatomy literature than any individual human could read in a lifetime. The footage doesn't lie. The anatomy doesn't change. The connection between what I'm observing and what the research names is either there or it isn't.
The people trying to use my openness about Claude as evidence that my work is fake don't understand the tool. They don't understand the process. And they have never explained what more rigorous verification would actually look like, because they can't. There isn't a better version of what I'm doing. There's just people who understand it and people who don't.
The environmental argument, and why it falls apart the moment you look at the numbers
This one is particularly interesting when it comes from people who are currently on TikTok.
The anti-AI environmental concern goes like this: AI uses a lot of energy, therefore using AI is environmentally irresponsible, therefore I'm a bad person for using Claude. It's tidy. It fits in a comment. It's also completely disconnected from what the actual numbers say, and I want to walk through those numbers carefully, because nobody making this argument has done that work.
A single AI text query, the kind I use to research anatomy, produces between 0.03 and 1.14 grams of CO₂e and consumes approximately 0.24 watt-hours of energy. For reference, the energy burned typing the prompt on your laptop exceeds the energy required to generate the answer. Google's research on the Gemini assistant found that the median text prompt is roughly equivalent to watching nine seconds of television. A thousand text queries can use as little as 9% of a full smartphone charge.
Now let's look at the platform the criticism is arriving from.
A single minute on TikTok generates approximately 2.921 grams of CO₂e, according to carbon accounting firm Greenly's analysis. The average user spends 95 minutes on the platform daily and opens the app 19 times. That's an estimated annual carbon footprint of 48.49 kg of CO₂e per user, equivalent to driving a gas-powered car 123 miles, every year, just from TikTok. TikTok's global carbon footprint is estimated at 50 million tonnes of CO₂ annually, a figure comparable to the total national emissions of Greece.
The reason is structural. Video streaming is the most energy-intensive activity on the internet. Text-based AI queries are among the least. TikTok's data centers still run primarily on conventional energy, coal and natural gas, with only one renewable-powered facility currently operational. Carbon accounting professionals have been direct about the implications: "The relationship between doomscrolling and carbon emissions is quite direct. Every minute spent on these platforms requires server processing power and data transmission. What many need to realize is that mindless scrolling has a very real environmental cost."
The person typing "your Claude use is bad for the environment" into TikTok, from TikTok, after ninety-five minutes on TikTok, is not making an environmental argument. They're pattern-matching to a headline they half-read and applying it to work they don't understand. They don't know what Claude is. They don't know what a text query costs. They haven't looked at their own numbers. And they're directing the criticism at the person doing the most targeted, low-footprint research use imaginable while sitting inside the second most carbon-intensive social platform on earth.
I'm not here to shame anyone's screen time. But I am done accepting environmental criticism that hasn't survived contact with basic math.
The actual sustainability story, the one nobody's telling
Now here's what I actually want to talk about. Because this is the part that reframes everything.
What does this work do in the world?
It shows people that the thing they were told they needed, the serum, the device, the filler, the procedure, is addressing a symptom of something that can be worked on internally. It gives people a framework for understanding their own anatomy instead of handing it over to someone else to fix. It keeps people out of consultations that turn into procedures they didn't need. It redirects money that would have gone to the cosmetic industry back into people's own understanding of their own bodies.
The global cosmetic surgery market was valued at over $50 billion in 2023 and keeps growing, driven substantially by the exact kind of appearance dissatisfaction that my work is directly countering. Every person who works with their fascia instead of booking a filler consultation is a real, measurable outcome. That's money not spent on anesthesia, on recovery, on the psychological cycle of chasing an external fix for something internal. That's a person who now has a tool they own, that costs nothing, that they can return to for the rest of their life.
That is a sustainability argument. A real one. About resource consumption, about the beauty industry's supply chain, about what it means to give people access to something that reduces their dependency on products and procedures they were being sold because nobody gave them another option.
So when someone sits in my comments to tell me that my use of Claude, a text-based research tool producing less carbon per query than nine seconds of televisionm is the environmental problem, while I am actively producing content that prevents people from financing cosmetic surgeries, I want to ask a sincere question.
What are you doing?
Not rhetorically. Actually. What is the net impact of the time you spent typing that comment? What did it prevent? What did it redirect? What did it give someone that they didn't have before?
Because I can answer those questions about my work. Every day, in my comments, from people who say they canceled a consultation. From people who finally understand why their face looks the way it does. From people who found something they can do themselves, for free, with their own hands, that actually changes something.
That is the sustainability story. That is what this work is for. And the people trying to discredit it with a half-formed argument about AI energy usage have not looked, at the numbers, at the tool, at the process, or at what I'm actually building here.
They don't know what I'm doing. They just know they don't like that it's working.
I found this in my body. I verified it through footage. I named it through research. I translated it into language precise enough that someone on the other side of the world can feel it in their own body.
That is the work. All of it. And every part of it matters.
Less carbon than your morning scroll. Fewer people financing cosmetic surgery.
That's what this actually does. And I'm just getting started.
