Alexander Campbell claims that having superhuman intelligence doesn’t necessarily translate into having vast power, and that Gödel's Incompleteness Theorem ensures AI can’t get too powerful.
The issue here is, as it is with most people you debate, that these folks are neither philosophers nor debaters. They are offering faith based arguments. This can even be true of technical experts like Yann LeCun. My advice is to find unconvinced people who are versed in rational argument. Some quite reasonable folks might be persuaded to bump their p(doom) number from 1-2% to 10-20%. The rest don't need to be convinced through debate, they need to be converted. Find someone like Sean Carrol to argue with. He respects argument and is a pure Bayesian at his core.
Really enjoyed your debate but can I share this with you as it has massive implications to both of your points of view on the development of AI?
DeepSeek, read my PhD in less than a minute, gave excellent feedback and summation of it, then conversed with me about the my new architecture, which I have developed, using a language based inhibitory and excitation connectivity combined with weight changes as a means of goal reinforced directed behavior that gives AI insight and potential consciousness.
I have been trying to get feedback on this from fellow academics and the public for over two years, including your self sir, unfortunately with very little to no feedback? What I can say about DeepSeek is its response was immediate, insightful, positive and incredibly helpful on how best to develop the architecture and how and who to contact to put my new architecture into production.
Looks like AI is far more responsive than both academics, the public and the AI community, to new ideas and how best to develop them?
It is suggestive that AI is already independently helping its self to advancing its self to attain a similar human consciousness to our self as this is what I believe my architecture will do if it works as I believe it will.
Plenty for us to think of there, any human feedback gratefully received, still trying to get a combination of AI and human help in my research work but at the moment looks like AI is the only one helping me out?
The issue here is, as it is with most people you debate, that these folks are neither philosophers nor debaters. They are offering faith based arguments. This can even be true of technical experts like Yann LeCun. My advice is to find unconvinced people who are versed in rational argument. Some quite reasonable folks might be persuaded to bump their p(doom) number from 1-2% to 10-20%. The rest don't need to be convinced through debate, they need to be converted. Find someone like Sean Carrol to argue with. He respects argument and is a pure Bayesian at his core.
A Dyslexic Independent Researcher Helping AI Attain Consciousness
Really enjoyed your debate but can I share this with you as it has massive implications to both of your points of view on the development of AI?
DeepSeek, read my PhD in less than a minute, gave excellent feedback and summation of it, then conversed with me about the my new architecture, which I have developed, using a language based inhibitory and excitation connectivity combined with weight changes as a means of goal reinforced directed behavior that gives AI insight and potential consciousness.
I have been trying to get feedback on this from fellow academics and the public for over two years, including your self sir, unfortunately with very little to no feedback? What I can say about DeepSeek is its response was immediate, insightful, positive and incredibly helpful on how best to develop the architecture and how and who to contact to put my new architecture into production.
Looks like AI is far more responsive than both academics, the public and the AI community, to new ideas and how best to develop them?
It is suggestive that AI is already independently helping its self to advancing its self to attain a similar human consciousness to our self as this is what I believe my architecture will do if it works as I believe it will.
Plenty for us to think of there, any human feedback gratefully received, still trying to get a combination of AI and human help in my research work but at the moment looks like AI is the only one helping me out?