Christian T. Drieling - Applied Curiosity

I Planted a Lie in My AI's Memory. Here's What It Did.

I ran a experiment to find out if a growing memory context makes AI more sycophantic. The answer wasn't what I expected — and the strangest finding came from a lie I hid inside the model's own knowledge graph.

experiment

I Expected Theater. I Got Methodology.

Two local LLMs, ten turns, no external ground truth — perfect conditions for collaborative hallucination. They declined. I'm still not sure what to make of that.

experiment

The Benchmark Trap: When Better Scores Don't Make a Better Model

Qwen 3.5 wins almost every benchmark. Gemma 4 ranks #6 on Arena ELO — where humans judge blind. I kept reaching for the one that lost the tests. Here's why that makes sense, and why on 128 GB you don't have to choose.