LLM benchmark and leaderboard for narrator-bias sycophancy, opposite-narrator contradictions, and judgment consistency.

Community-driven behavioral reliability benchmark for LLMs. 231 probes across 19 modules, deterministic scoring, perplexity correlation, layer sensitivity mapping, quant method capture, hardware-stratified community rankings. Every test contributes to the community dataset.

👟 SUP: Sycophancy Under Pressure

80,433-trial study of context-window sycophancy across 6 LLMs (4B–72B). Behavioral ratchet effect, correction injection mitigation, phase transition analysis. Code, data, and preprint included.

A sycophantic tool for preventing worse sycophancy.

Adversarial testing of LLMs on constraint satisfaction deadlocks

Falsification Chain of Thought — post-hoc verification of AI judgments using falsificationism to counter sycophancy bias

This repo shows the coding of sycophancy in LLMs as Bayesian-Latent model

Which LLM do you actually trust? Blind-test 100+ AI models with truth scoring and reasoning failure classification. No branding, no marketing — just data.

Anti-sycophancy memory layer for LLMs — models the relationship, not just facts. Gives the AI confidence to push back when you need it.

Systematic probing toolkit for alignment-relevant LLM behaviors: sycophancy, sandbagging, power-seeking, deceptive alignment, and corrigibility failures

Three-layer sycophancy defense skill for Claude Code and OpenClaw, based on ArXiv 2602.23971

AI with clerk discipline. Paste one prompt into any AI chat and every factual claim gets tagged by confidence. No install.

Official code for "From Fact to Judgment: Investigating the Impact of Task Framing on LLM Conviction in Dialogue Systems" (IWSDS 2026)

Prompt prevents contextual bleed between active discussion and meta object being discussed via conceptual scaffolding.

Benchmarking software engineering sycophancy (extreme agreement) in frontier LLMs

Empirical benchmark testing whether LLMs abandon correct answers under social pressure

Open-source benchmark that measures AI sycophancy: how often LLMs abandon correct answers when users push back. Tests GPT-4, Claude, Gemini & more across 50 questions using 5 pressure strategies. Produces a ranked leaderboard with capitulation scores.

A deterministic safety layer for probabilistic AI systems — preventing delusion reinforcement and AI-induced psychological harm through immutable governance