Summary of the Core Content
This article discusses the “miracles” and risks of AI in the medical field: It begins with several touching examples of how AI has helped in healthcare (such as using an AI communication app to help an autistic child speak, and using AI to develop a vaccine to save a cancer-stricken dog), but then points out that the role of AI in these stories is often exaggerated. Behind these stories lies the logic of entrepreneurs seeking attention, AI companies using them as marketing tools, and capital chasing trends. The article draws parallels with the “scam of vitamin C” from Silicon Valley’s history, warning of a potential new wave of false prosperity and deception in the field of AI-powered medicine, and urges the public to approach these claims with skepticism.
Detailed Analysis
1. What are those widely shared AI healthcare “miracles”?
Recent AI healthcare stories have given people the impression that technology can truly change lives:
- The communication app for the autistic child: A foreign engineer used AI to generate images related to his son and clone his voice, creating an animated app that helped the child, who had not spoken in four years, say “I really love you.” The child’s ability to communicate doubled within two weeks.
- AI saving the cancer dog: Paul, an Australian programmer, used GPT-4 and AlphaFold to identify a specific marker for cancer cells and collaborated with a university to develop an mRNA vaccine, which reduced the tumor by 75%.
- The Alzheimer’s drug developed in a garage: Douglas Yao, a Harvard-educated doctor, used his personal savings to develop PAC-832, the world’s first highly selective GalR1 antagonist, at a fraction of the cost of traditional research methods.
- AI helping diagnose lymphoma: An American entrepreneur used AI to analyze his blood data and symptoms, leading to an accurate diagnosis of non-Hodgkin’s lymphoma and avoiding unnecessary radiation therapy.
These stories suggest that technology can be a powerful tool for medical breakthroughs.
2. Is AI really capable of curing diseases? Or is it just a “tool”?
The role of AI in these cases has been greatly exaggerated:
- The vaccine for the dog: While Rosie received the AI-developed mRNA vaccine, she was also treated with two other cancer drugs; it’s unclear which one played the key role.
- The drug developed in the garage: AI helped in organizing information and identifying potential targets, but the actual research and development relied on the doctor’s expertise, and the drug is still in the early stages of testing.
- The autism app: While AI generated helpful images and synthesized the child’s voice, the child’s progress was largely due to the parents’ dedication (using a familiar animation style and the parent’s own voice). In essence, AI acts as a tool to assist professionals, not as a miracle-maker.
3. Why have these stories gone viral? Behind them are interests in traffic, capital, and the idea of “medical equality”
The widespread popularity of these stories is the result of a combination of factors:
- Entrepreneurs: They use them to attract attention and investment (e.g., Douglas Yao’s 16 tweets about his new drug, after transferring the patent to his company).
- AI companies: They use these stories to showcase their technology (e.g., GPT-4 and AlphaFold) and potentially make money by selling related products.
- The public: There is a desire for “medical equality”; advanced treatments like gene sequencing and custom vaccines were once only accessible to large pharmaceutical companies or the wealthy, but these stories give ordinary people the illusion that they can also benefit from AI in healthcare. However, most of the people involved in these projects (e.g., programmers and doctors) have relevant backgrounds.
4. Beware! The “exaggeration epidemic” from Silicon Valley is spreading to medicine
Silicon Valley’s tendency to exaggerate has spread to the medical industry:
- Historical lessons: The Theranos scandal showed how Elizabeth Holmes deceived investors with claims of a revolutionary blood test for cancer.
- Current examples: Companies like Medvi claim to provide remote healthcare services using AI, but their practices (fake patient data, fake doctors) expose them as scams.
- Large companies’ cheating: Meta manipulated model versions to rank higher in evaluations; OpenAI’s former developers pointed out that most AI companies are using similar techniques without genuine innovation.
Medicine is a field where lives depend on accurate information, and such exaggeration can lead to harmful scams.
5. What should the public do? Be cautious and wait for verification
When faced with claims of AI-powered medical miracles, we should:
- Avoid being swayed by emotions: While touching stories are inspiring, it’s important to verify the facts (e.g., whether other treatments are available and whether the technology has undergone clinical testing).
- Wait for verified results: Medical technologies must undergo rigorous trials; individual cases alone cannot prove their effectiveness.
- Be skeptical of quick success stories: Medical research is a lengthy process; claims of “instant vaccines” or “garage-made drugs” are likely exaggerations or scams.
In summary, while AI can indeed contribute to medical progress, it should not be regarded as a panacea. The difference between a miracle and a scam is often subtle. It’s best to wait for more evidence before accepting claims about the revolutionary potential of AI in healthcare.