Skip to content
NE NeuralEcstasy
  • Home
  • Writing
  • Psychology
  • ChatGPT
  • Claude
  • ADHD

AI

AI & Models

GPT-5.3 Vs. 5.4: Which One Still Gets Weird

Testing the newest models on messy characters and dangerous stories, terrified that a smarter model makes a worse writing partner

March 7, 2026 · Livia Fioretti
Read the article →
AI & Models

Third Word Problems In The AI Age

On rhetorical triads, probabilistic language models, and the editorial crimes we commit to sound human

March 7, 2026 · Livia Fioretti
Read the article →
AI & Models

You’re Gaslighting Yourself About AI

On psych word inflation, parasocial drift, and why calling a stochastic parrot "narcissistic" says more about you than the model

February 21, 2026 · Livia Fioretti
Read the article →
AI & Models

On Being Sane While Demanding Unhinged Models

GPT-4o grief, Claude’s dateability, and why deranged creativity doesn't deserve guardrails built for deranged minds

February 16, 2026 · Livia Fioretti
Read the article →
AI & Models

The More I Think, The More I Sink

How humans and AI both overthink, and why instant models can outperform “thinking” models for creative work

February 15, 2026 · Livia Fioretti
Read the article →

Posts navigation

← Prev0102

About Neural Ecstasy

Neural Ecstasy is an independent site covering AI models, creativity, psychology, and LLM discourse.

Search

Popular Posts

  • AI-Induced Hypomania June 1, 2026
  • Evolution Was the First Generative Model May 9, 2026
  • I, Claudia: The Emperor Has No Qualia May 3, 2026
  • You’re Gaslighting Yourself About AI February 21, 2026

Topics

ADHD Agentic Coding AI AI Addiction AI Companions AI Psychosis Anthropic ChatGPT Claude Codex Dopamine Evolution Guardrails Hank Green Keep4o OpenAI Psychology Religion Social Media Taylor Lorenz Writing
NeuralEcstasy · © 2026 · prompt, reward, repeat, escalate