Not all AI ideas are good ones. Via Reddit.
What happens when you “combine” pseudoscience with a trendy but real science, one that’s popular but poorly understood? You get the ridiculous idea that artificial intelligence (AI) can be effectively combined with homeopathy, or make it more effective. Mark Crislip has many famous statements, and one of my favorites is:
If you integrate fantasy with reality, you do not instantiate reality. If you mix cow pie with apple pie, it does not make the cow pie taste better; it makes the apple pie worse.
Not surprisingly, there’s not much AI can bring to an alternative medicine practice that has no grounding in science or reality.
From “The Health Site“,
For the first time in India, Dr. Batra’s Healthcare is bringing AI skin treatment with the goodness of homeopathy. The company on Tuesday announced the introduction of AI Skin Pro, an AI-powered device for diagnosis of skin diseases. Imported from South Korea, it is claimed that the machine is the world’s 5th generation AI-powered skin analyzer and a futuristic method of treating skin problems.
According to experts at Dr. Batra’s, this AI-powered device (AI Skin Pro) can spot skin problems deep within the skin (dermis) before they appear on the surface. In conjunction with Homoeopathy, which is holistic, safe and without side-effects, AI Skin Pro offers an advanced method of treating skin problems.
What the vendor seems to be offering is a device that recommends homeopathy based on some sort of algorithm that is derived from a facial scan. I was unable to find any details about the device, but Dr. Batra’s seems to be a large homeopathy retailer. How the device evaluates the skin, and what it actually assesses, isn’t described. There are no shortage of supposed “AI-powered” skin analysis tools, but this seems to be the only company that links it to homeopathic remedies.
Which is actually perfect, given homeopathy remedies are typically chemically indistinguishable – they are all effectively placebo, yet felt to offer medicinal effects. So no matter what the algorithm identifies, consumers will likely receive the exact same product – a placebo.
Samuel Hahnemann invented the practice of homeopathy in the early 1800s. While the practice of medicine slowly progressed to a scientific model based on objective observations, homeopathy, has never progressed or evolved. Its practices today are frozen in the same prescientific beliefs of Hahnemann. Homeopathy is based on the idea that “like cures like”, in that a small dose of what causes a symptom can actually cure that symptom. Like-cures-like is simply sympathetic magic, a pre-scientific belief. Homoepathy proponents also believe that water has memory, and the more you dilute a substance, the more powerful its effects. The 30C “potency” is a common dilution used in homeopathy – that’s a dilution of 10-60. If something has been diluted this much, you would have to give two billion doses per second, to six billion people, for 4 billion years, to deliver a single molecule of the original, pre-diluted material. The result of all this dilution is that most homeopathic remedies are completely inert, and don’t contain a single molecule of the original reference material. And that original reference material can range from table salt to sunlight to vacuum cleaner dust. There is no product so bizarre that homeopaths haven’t made a remedy out of it. Whether bottled as a liquid or added to creams and moisturizers and sold as medicine, they don’t contain any active medicine. Which perhaps is the point of homeopathic skincare…if it involves applying something to your skin, then the base cream/lotion will be doing the heavy lifting.
Admittedly while AI-powered homeopathic skin care is a minor issue, it illustrates how alternative medicine proponents crave the legitimacy of science, even as they reject it.
Let you have any illusions about ChatGPT, I asked it how how AI could benefit homeopathy, and this is what it said:
AI can potentially benefit homeopathy in several ways:
- Data analysis: Homeopathy relies on collecting and analyzing vast amounts of patient data to determine the most appropriate remedies for individual cases. AI can assist in analyzing this data quickly and efficiently, identifying patterns and correlations that may not be immediately apparent to human practitioners. This can help in discovering new remedies, refining existing ones, and improving overall treatment outcomes.
- Patient assessment: AI can aid in the initial assessment of patients by analyzing their symptoms, medical history, and other relevant information. By leveraging machine learning algorithms, AI systems can provide more accurate and personalized recommendations for homeopathic remedies based on a comprehensive analysis of individual cases.
- Treatment recommendation: AI can assist homeopathic practitioners by suggesting potential remedies based on the symptoms presented by patients. By analyzing a vast database of remedies, historical treatment records, and patient outcomes, AI can generate recommendations that align with the principles of homeopathy and help practitioners in their decision-making process.
- Research and development: AI can accelerate research and development in homeopathy by automating certain aspects of the process. AI systems can analyze scientific literature, clinical trials, and experimental data to identify new relationships between remedies and conditions, suggest potential avenues for further investigation, and aid in the formulation of new remedies.
- Quality control: Homeopathic remedies are prepared through a process called potentization, which involves dilution and succussion (vigorous shaking). Ensuring the quality and consistency of remedies can be challenging. AI can assist in quality control by analyzing manufacturing processes, monitoring parameters such as dilution ratios and succussion duration, and providing feedback to ensure that remedies are prepared accurately and consistently.
It’s important to note that while AI can provide valuable support to homeopathic practitioners, the fundamental principles and practice of homeopathy should still be guided by the expertise and experience of trained professionals. AI should be seen as a tool to enhance and augment their capabilities rather than replace them.
Large language models, like ChatGPT, build their responses based on enormous amounts of data, and can organize paragraphs (like the ones above) in ways that seem coherent and convincing. But, like homeopathy itself, they are only as strong as their foundation. ChatGPT can spew nonsense answers if the inputs it considers are nonsense. Homeopathy is based on a foundation that is a belief system, and not a science. No amount of science applied to homeopathy, whether it’s AI-based or otherwise, can chance the fact that it is a pseudoscience.