Skip to content

R0028/2026-03-26/C008

Claim: Approximately 84% of recommendations in the official prompt engineering documentation from OpenAI, Anthropic, Google, and Microsoft are subjective or qualitative, with only about four out of roughly 25 distinct recommendations including any quantifiable criteria.

BLUF: Cannot be independently verified from web search alone. This claim requires a systematic content analysis of the official prompt engineering guides. The guides from these four companies are primarily qualitative in nature (e.g., 'be specific,' 'use examples,' 'break tasks down'), and the overall characterization is plausible, but the specific 84% figure and 4-out-of-25 count require original analysis not available in published literature.

Probability: Roughly even chance (45-55%) | Confidence: Low

Correction needed: The specific percentages (84%, 4/25) could not be independently verified. The qualitative characterization — that most recommendations are subjective — is consistent with the documented content of these guides.


Summary

Entity Description
Claim Definition Claim text, scope, status
Assessment Full analytical product with reasoning chain
ACH Matrix Evidence x hypotheses diagnosticity analysis
Self-Audit ROBIS-adapted 4-domain process audit

Hypotheses

ID Hypothesis Status
H1 Claim is accurate — exactly 84% are subjective Inconclusive
H2 The characterization is directionally correct but the specific percentages are unverifiable Supported
H3 Claim is materially wrong Eliminated

Searches

ID Target Results Selected
S01 Primary search 10 3

Sources

Source Description Reliability Relevance
SRC01 OpenAI, Anthropic, Google, Microsoft prompt engineering guides High High