Future of Science and Technology Q&A (February 14, 2025)

The Stephen Wolfram Podcast

Stephen Wolfram answers questions from his viewers about the future of science and technology as part of an unscripted livestream series, also available on YouTube here: https://wolfr.am/youtube-sw-qa

Questions include: In the spirit of Valentine's Day, what is the future of bionic hearts? Would this be a way to make humans more efficient? - How would you think about a world in which all of the work is done by robots and AI? - Architects using computerized/AI tools will result in less demand for architects overall, thus less people getting to do architecture. - Do you think the current methods of training and using AI/LLM are here to stay for a while, or is there a real possibility of an alternative machine learning approach appearing and being superior and more efficient? - What would you think about spiking neural nets with a new non-differentiable learning algorithm? Is it the path to smarter AI? - ​​I read from an expert that correcting the errors in a later prompt results in more errors. It's better to go back to the original prompt. - Do you or your team actively work on the alignment issues with AI and are you worried about the next 10 years with regard to that? - Do you see a danger in the trend toward anthropomorphic AI and providing AI systems with human-like attributes? - I wonder what will happen when future AI models are trained on material that shows them the actual results of their past actions. - What are some near- and medium-term breakthroughs that could potentially make the creator a trillionaire? Off the top of my head, fusion power, far more efficient batteries or novel propulsion systems.