Space Cat's Zero-G Jump in Str...
We are living a grand illusion. As we watch Large Language Models (LLMs) draft essays, write functional code, and converse with poetic nuance, we operate under a quiet, dangerous assumption: that if we just stack enough parameters, compute cycles, and training text high enough, true understanding and intent will eventually spark at the top.
It won’t.
#programming #AI #intelligence #cogsci #philpsy #neurosky #philosophy

We are living a grand illusion. As we watch Large Language Models (LLMs) draft essays, write functional code, and converse with poetic nuance, we operate under a quiet, dangerous assumption: that if we just stack enough parameters, compute cycles, and training text high enough, true understanding and intent will eventually
I'm at #Duke to share methods and results from Socratic Reflection experiments to the Advances in Decision Analysis #conference:
#2026ADA website: https://conferences.fuqua.duke.edu/2026-informs-advances-in-decision-analysis-conference
Follow the URL below for our results and my posts about other talks or posters I see.
Can 3-6 year-olds distinguish between reliable vs. unreliable informants?
A study of 93 #kids found the children were sensitive to source reliability, with limited abilities to generalize across contexts.
This notebook explores "epistemic curvature" and homotopy-theoretic #activeinference using a REAL WORLD example.
The example is very simple, but there are plans to extend it to large-scale social networks in the future.
https://colab.research.google.com/drive/16tUQiybglvn2RfzKuu_0uoI_drPQIRYS?usp=sharing
Why can’t the tech world agree on whether AI actually "reasons"?
It turns out, the answer is hidden in how we talk about feeling cold.
https://www.ocrampal.com/i-am-vs-i-have-how-everyday-grammar-secretly-influences-the-ai-debate/
#programming #AI #intelligence #cogsci #philpsy #neurosky #philosophy

When you say "I am cold," you aren't just describing your temperature. You are picking a side in a centuries-old philosophical war, one that perfectly explains why the tech world is currently tearing itself apart over whether AI can truly "reason." There is a curious linguistic asymmetry hiding in plain
Do logical visual aids (such as Venn diagrams) improve argument evaluation?
Not in this study of 164 English speakers (with "normal or corrected-to-normal vision and no issues seeing color").
https://doi.org/10.1111/cgf.70476
#logic #teaching #edu #visualization #CogSci #learningScience