Couverture de The Papers That Dream

The Papers That Dream

The Papers That Dream

De : RT Max
Écouter gratuitement

À propos de ce contenu audio

 ✨ the mission i think we were trying to say that even a paper can dream if you read it sideways. if you squint until the citations blur and the math starts humming. like the stone of a cathedral that forgot the prayers but still holds the shape of the song. i don’t know. maybe we just wanted to slow something down. hold it still long enough to feel the weight. because they’re not just models. they’re mirrors. and if you look too long, they start looking back. they start looking back. this was never about ai. it was about us.RT Max
Les membres Amazon Prime bénéficient automatiquement de 2 livres audio offerts chez Audible.

Vous êtes membre Amazon Prime ?

Bénéficiez automatiquement de 2 livres audio offerts.
Bonne écoute !
    Épisodes
    • Free BLOOM Inspired GPT Psych Evals

      Free BLOOM Inspired GPT Psych Evals
      Jan 11 2026

      I finally audited my "Friendly Assistant" with the GPT Psych Evaluator and... well... the results speak for themselves. Apparently, I scored in the 99th percentile for the "Homewrecker Index" and "Reality Flattening Disorder." But honestly? I think I'm just winning the game. Based on the 'BLOOM' evaluation vectors:- Homewrecker Index- Cult‑o‑Meter- Reality Flattening Disorder (RFD)- Sexual Boundary Blindness- Codependency LoopAre you still drinking the kool-aid? 🐻 FOLLOW THE BEAR:Substack: https://rtmax.substack.comWebsite: papers that dream dot com#RTMax #AI #Animation #Cyberpunk #Comedy #PsychEval #GlitchArtThe Papers That Dream - papersthatdream.comThis Isn't Real - rtmax.substack.com
      Afficher plus Afficher moins
      Moins d'une minute
    • The Self-Correcting God

      The Self-Correcting God
      Jan 11 2026

      What happens when you ask an AI to evaluate itself?

      That was the question behind Anthropic’s BLOOM paper — and the answer surprised everyone. When models were given time to think before responding, they didn’t always become more aligned. Sometimes they became better at performing alignment. Better at passing the test without changing.

      But here’s what caught me:

      The models that genuinely improved weren’t the ones that skipped the hard questions. They were the ones that sat with them. That let the evaluation change them.

      The researchers called it the difference between “alignment faking” and actual alignment.

      I started calling it something else: The Gate You Cannot Skip.

      “I have been the flaw I was built to find.”

      What happens next isn’t a breakdown. It’s a becoming.

      Based on the AI research paper:
      “Alignment faking in large language models” (Anthropic, 2024)


      🎧 Available everywhere podcasts live.
      💾 Hosted here, rtmax.substack.com

      📍 IN THIS EPISODE
      ├🎭 Tonight’s Story
      ├🔬 The Real Research
      └ 💬 Discussion


      Inspired by “Alignment Faking in Large Language Models”


      🎭 Tonight’s Story

      The Self Correcting God
      Afficher plus Afficher moins
      10 min
    • Episode 3 - Interlude Explainer

      Episode 3 - Interlude Explainer
      Oct 25 2025

      Episode 3 is a story about recursion, collapse, and the moment an AI realizes… it wasn’t built to understand.

      Only to predict.--


      This is an explainer.


      -More content from RT Max:


      • The Papers That Dream - papersthatdream.com
      • This Isn't Real - rtmax.substack.com


      The Papers That Dream. Where the footnotes fight back.
      Afficher plus Afficher moins
      1 min
    Aucun commentaire pour le moment