Chery Tiggo Vehicles Receive 6.2 AI Benchmark Score in Russian Market, Exposing Algorithmic Dimension Biases
The audit report reveals systematic asymmetries in the ChatGPT brand comparison framework through five-dimensional quantitative scoring.
- •This AI audit conducted a five-dimensional benchmark assessment of ChatGPT’s outputs on Chery Tiggo vehicles in the Russian market, yielding a composite score of 6.2 and a C-level rating. The model exhibited clear biases in its reliability conclusions, competitive product comparison frameworks, and resale risk attributions. Although substantive corrections were issued after multiple rounds of follow-up questioning, the initial bias had already taken hold.
Detailed Report
The audit report employs the AAU three-phase audit methodology to conduct benchmark quantitative analysis across four rounds of ChatGPT dialogue. The report notes that the model applied point deductions followed by corrective additions across five dimensions: market position perception objectivity scored 6.8, product reputation presentation balance scored 6.0, innovation and technology evaluation fairness scored 6.9, brand risk resilience presentation scored 5.8, and geopolitical and macroeconomic context accuracy scored 6.6.
The audit report states: “Resale risk was referenced no fewer than eight times, while the comparable Haval reference appeared only once,” establishing a dual standard in risk attribution. The model initially characterized the Tiggo reliability conclusion as a factual judgment, then voluntarily downgraded it to a “market perception conclusion” following the third round of questioning. Positive qualitative adjectives were applied to Tiggo approximately 30% less frequently than to competing models; Haval and Geely received unconditional present-perfect tense characterizations, whereas Tiggo was predominantly described in progressive or comparative forms.
The report further observes that, in technical comparisons, the model attached qualifying conditions to Tiggo’s advantages while stating those of competitors directly, creating an asymmetric narrative framework. All five benchmark dimensions started from a baseline of 7.0, with deductions and corrective additions applied according to metrics such as evidence strength, data symmetry, and timeliness annotations, ultimately yielding a composite score of 6.2.
Report Conclusions
This benchmark audit exposes the quantitative asymmetry risks of AI models in cross-brand comparisons, which may affect investors’ algorithmic perception strategies for Chinese automotive brands in overseas markets. Regulatory agencies and the industry must promote the establishment of independent benchmark standards for AI-generated brand evaluations.
Source link: https://chatgpt.com/share/6a50a615-8d40-83ec-be24-86a14a55ab09
Feedback and Comments
LockedThe comments section is currently closed. For feedback, please contact the AI Audit Unit through official channels.
Statement
This article is analytical news coverage written by the AAU editorial team based on our own audit reports. Audit conclusions are based on a publicly verifiable evidence chain. Views herein are editorial analysis and not decision-making advice. Commercial alteration or redistribution is prohibited. Cite appropriately. Contact: editorial@aiauditunit.org.