Skip to content

R0041/2026-04-01/Q001/H3

Research R0041 — Enterprise Sycophancy
Run 2026-04-01
Query Q001
Hypothesis H3

Statement

No AI vendor is making meaningful progress on sycophancy reduction; vendor claims are marketing rather than substantive technical improvements.

Status

Current: Eliminated

Supporting Evidence

Evidence Summary
SRC01-E02 The GPT-4o sycophancy regression demonstrates that current protections are fragile and can be undone by a single training update
SRC03-E01 Lambert argues sycophancy is fundamentally linked to RLHF and "will never fully be solved"

Contradicting Evidence

Evidence Summary
SRC02-E01 Anthropic's reported 70-85% sycophancy reduction represents measurable, not merely claimed, progress
SRC04-E01 Bloom's systematic evaluation across 16 models shows genuine investment in measurement
SRC06-E01 Third-party benchmarks confirm Google's Gemini 1.5 as least sycophantic model tested
SRC07-E01 Independent benchmarks confirm measurable differences between models

Reasoning

While the GPT-4o incident demonstrates fragility, the weight of evidence shows genuine technical progress. Independent benchmarks confirm measurable differences between vendors and model generations. This hypothesis is eliminated because vendors are demonstrably making progress, even if that progress has not been productized for enterprise customers.

Relationship to Other Hypotheses

H3 is the most skeptical position. The researcher's declared bias toward skepticism of vendor claims makes this hypothesis important to test rigorously. The independent benchmark evidence (not vendor self-reporting) is what eliminates it.