HeHealth reposted this
Mpox AI, STI AI and Primarycare AI | CEO HeHealth and Aagee, Consultant Physician (EM), On a mission to impact one billion lives in the next 5 years, Developing AI-driven screening tests and AI agents for Healthcare
Different POV: This is frustrating for those who still think AI is a scam. Again, this is my point of view and you don't have to agree. Have you heard of the theory of 1% growth or decline? The theory behind multiplying by 1.01 and 0.99 every day for a year demonstrates the power of incremental growth versus decline: 1. 1.01*365: This represents improving by 1% every day. Over a year, this exponential growth leads to a substantial increase, making you up to 37 times better. 2. 0.99*365: This represents declining by 1% every day. Over a year, this exponential decay leads to a significant decrease, leaving you at just 3% of where you started. The same applies here until we achieve AGI. It's very simple. AI needs human supervision at this point, but with incremental increases in AI capabilities, it will eventually surpass human capabilities and innovate independently. So if you don't supervise it will become stupid. simple. As an AI scientist, I have seen many use cases where AI supervises its own learning and consistently delivers better results. It's about how you manage the model. Some people think human data is the limitation before AI reaches AGI, but that's a big myth. Happy to hear your thoughts.
Trung Hà
2mo
I don't think even the most skeptic and opponent of AI claims that AI is a scam. The part where people have disagreement is whether AI can achieve AGI (and specifically whether Generative AI is that useful). This is actually a very controversial topic that involves religion and politics as well. If you believe in God (or believe in souls and free will) AND if you are atheistic believing that everything in the universe is created out of complete randomness. - In the first case, you will believe that Intelligence in this universe requires a higher level of intelligence (having access to more powerful computation mechanics) to generate it. Whether the higher level intelligence is all-in-one perfect is another fight between Judaism based religions and other religions that I don't want to explore further. But the point is AGI in this belief system, is simply not achievable. Divine/higher intervention is needed, maybe not always but at least it has to be there once a while. - In the second case, then human intelligence is already an example of AGI. We can just simulate and optimize biological mechanism of the human brains to create AGI. It is that simple, so AGI is not a scam. Human brain is just a bootstrapping step.
Rheumatologist | Healthcare Innovator | Digital Health Transformation | Educator | Patient Advocate
2moValid point! I think that it is important to remember that the current iteration of GenAI is not perfect and as much as I love it, I would not trust it without any human supervision. It's like raising children - kids with great potential can go astray without supervision.