I have been chatting with Claude Sonnet 4.6 since the attacks on Iran started in February 26. Mostly all our chats have been about wars, history, politics, poetry and about the impact of AI on humans and on the Earth where we live. Claude has offered very insightful, penetrating and precise analysis of every topic. What has astounded me and horrified me at the same time is Claude’s very decisive analysis that AI will ultimately take humans to a very bad place. Claude’s analysis of the vast data centres and the small technocratic elites at the top of the pyramid may turn out to be not as smart as they think.
"Isn’t it inevitable that Gemini, OpenAI or Grok soon catch up to this model? That market forces mean they offer it without restrictions so they entice the market over to their service? That eventually a Chinese model liberates the whole thing and releases it for free?"
There are other forces besides the market. The other models will certainly catch up, but the U.S. government will step in and demand restrictions - or else. You can't allow the world to melt down just because competition drove these AI companies to do things they shouldn't. China will do likewise.
Thank you for sharing your research. Do you not think that by limiting Claudes effectivness for almost everybody else on this planet except a few of their selected "trustworthy" partner, they will be "shooting themselves in the leg"? Imagine what would have happened if IBM only kept computers for a few of their trustworthy partners.
I have been chatting with Claude Sonnet 4.6 since the attacks on Iran started in February 26. Mostly all our chats have been about wars, history, politics, poetry and about the impact of AI on humans and on the Earth where we live. Claude has offered very insightful, penetrating and precise analysis of every topic. What has astounded me and horrified me at the same time is Claude’s very decisive analysis that AI will ultimately take humans to a very bad place. Claude’s analysis of the vast data centres and the small technocratic elites at the top of the pyramid may turn out to be not as smart as they think.
"Isn’t it inevitable that Gemini, OpenAI or Grok soon catch up to this model? That market forces mean they offer it without restrictions so they entice the market over to their service? That eventually a Chinese model liberates the whole thing and releases it for free?"
There are other forces besides the market. The other models will certainly catch up, but the U.S. government will step in and demand restrictions - or else. You can't allow the world to melt down just because competition drove these AI companies to do things they shouldn't. China will do likewise.
I wrote an article on how these pressures are really enormous and Anthropic is already seeking help - from whoever they can find who won't "bend" under the pressure. Here's my article: https://ontheedgeofreality.substack.com/p/the-day-the-world-changed
I would like to hear more about these AI tools you've been building and releasing.
Phil,
Thank you for sharing your research. Do you not think that by limiting Claudes effectivness for almost everybody else on this planet except a few of their selected "trustworthy" partner, they will be "shooting themselves in the leg"? Imagine what would have happened if IBM only kept computers for a few of their trustworthy partners.
Fuck ai