Matt Going back to you not being able to locate my research, and concluding I was a fake researcher. Based on my investigation, your perspective was valid, but unreliable and publically damaging.
I since discovered that researchgate "cancelled" my public profile without ever raising a query with me. However, when I logged in, all my research would always display. Even AI could not locate my research, or existing citations. Some of those papers were peer reviewed. I pointed out to the publishers that a cascade delete would cause tremendous damage to associated, published papers. I confronted their sys admin and while they feigned confusion, my public profile suddenly reappeared and significant activity ensued. It's been less than a month, and I noticed how again activity on my profile plummeted, even though new research was published. Not even a single view. Nah! This is the same pattern as before. So, the human-driven algorithms seem to be sanctioning novel and impacting research. I take this as a win. This seems to ve part of an emerging "research harvesting" AI-driven culture. Confirmed to me as a growing trend by AI. It could also be untouchable humans in organizations controlling the scientific narrative while misappropriating research for own gain. All my tech's my own and the latest are under public, commercial license protection. Not that it protected my work before, but one has to persevere. Independent researchers are sitting ducks. Less than 1.5% of 0.00001% of the global pool of novel researchers ever attain publication. That translates into 2 persons for the potential 2400 - 8200 researchers of novel science out there. I've submitted my latest novel research to a mathematics journal. After about a month, I'm still struggling to get them to acknowledge receipt of my research, let alone a review decision. Soon though, I'll secure more-appropriate interest abroad and withdraw my submission. Meantime, the preprint serves as prior art. Seems, the real issue with researchgate was with the cov epidemiological systems research, which I published a while ago with full sources listed and using an IEEE-published method set. While no one challenged, AI discovered a vague metatag suggestion that "either my research was removed or I was spreading false info". AI challenged my resesrcher claims as well, since rectified. The constructivist research must've hit a raw nerve somewhere. All my research is falsifiable, but no party ever bothered. I'm validating your concerns. I apologize without reserve for having been at loggerheads with you and for insulting you in any way. I'm now systematically relocating my research to a reliable publisher. My research, my problem. On Fri, 20 Mar 2026, 15:48 Matt Mahoney, <[email protected]> wrote: > The article says that software developers using GitHub Copilot spend more > time on code development and less time on project management. About 96% of > developers don't trust AI generated code, so they spend more time reviewing > and fixing it, which takes more effort than talking to the human who wrote > it. > > https://www.reddit.com/r/ArtificialInteligence/s/bpjIpTbtrk > > In order words, AI has the same effect on developers as everyone else. We > spend more time talking to machines and less time talking to people. > > > -- Matt Mahoney, [email protected] > *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/T6540dc690a896d83-Mf4b3058801e06691181ee705> > ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T6540dc690a896d83-M29d66d7386fae5d523588ee6 Delivery options: https://agi.topicbox.com/groups/agi/subscription
