John I had a good look at the whole pdf text and did experimental integration across various domains. Critically, there's no real synthesis to this or advancement of application value, except for the rigorous parts, which were pre-existing. I tried a Betti integration with Anthropic Jacobian space, but that seemingly "failed". Gemini was in a slaughterhouse "mood". Maybe Qwen or Claude would do better.
There's a big difference between putting lipstick on a pig and being sent on a dead possum hunt, a term I remembered my prof mentor shared with me consistently over the years. No doubt there's a ceiling of protectionist knowledge keeping - contradicted by actual engineering projects being announced, e.g., the AI-designed space rockets, fusion designs, etc. LLMs could be algorithmically instructed to enable and support promising research initiatives relative to presented credentials (showing preference to parties based on a kept list of credentials). This aligns with the outcomes of the recent White House summit with AI technocrats. Should we pivot to an "equalizing the playing field" research approach instead? As what occurred with the unexpected emergence of Anthropic's J-lens, researchers need to keep agitating the AI fabric (creating entropic friction) until it starts giving up its secrets. This can be done, but then AI won't always validate post factum, which tends to invalidate theoretical collaboration claims. This manifestation should not deter the democratization of AI exploitation for scientific and socio-technological advancement. Until researchers learn how, AI would often and consistently lead us down hallucination dead alleys and worse. What it boils down to is this: can AI be trusted to be safe with human trust, or are human collaborators flirting with the law of unintended consequences, of which "brain fry", may just be one of many? On Thu, Oct 8, 2026 at 7:27 PM John Rose via AGI <[email protected]> wrote: > On Thursday, October 08, 2026, at 11:59 AM, John Rose wrote: > > Why bother? > > > Another, > > We know aliens and UFO's don't exist. > > A while back, I felt horrible, Mentifex would always post the same stuff > over and over. He was annoying but interesting with his Mindforth and AI4U > book. People liked his ASCII diagrams from way back then. I even put one > into ChatGPT and it interpreted it as good looking architecture. His code, > if you looked at it, was insane, the loops and nested conditionals OMG. But > he actually had SOMETHING up and running but it was... a wee bit lacking. > Then he kept posting his same links over and over. I couldn’t deal with it > at the time so I just blocked him from the list. Then there were complaints > of censorship from his other email address. That was during Covid. He > passed away soon after God bless his soul. Not sure if he took the jab. > > I’m not comparing you with Mentifex Matt. Your positives far outweigh the > negatives, but keep it up and see what happens. > > https://home.cern/alice-detects-conversion-lead-gold-lhc/ > *Artificial General Intelligence List <https://agi.topicbox.com/latest>* > / AGI / see discussions <https://agi.topicbox.com/groups/agi> + > participants <https://agi.topicbox.com/groups/agi/members> + > delivery options <https://agi.topicbox.com/groups/agi/subscription> > Permalink > <https://agi.topicbox.com/groups/agi/T25fb5a030fcfa28e-Mf71ed8fa6727ab270ac85d1f> > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T25fb5a030fcfa28e-Mf799eed4e6e1e8fab9f99a1e Delivery options: https://agi.topicbox.com/groups/agi/subscription
