Summarize these selected article findings and their source context. This brief is not the full article. Preserve units, populations, dates, assumptions and limitations. Distinguish illustrative calculations, vendor descriptions and measured results. Do not invent omitted information or claim to have read the complete article. AI shopping recommendation accuracy statistics Selected article findings, not the complete article. Finding the product and answering the buyer’s question are separate tests. In the 2026 EComAgentBench study of 662 shopping tasks, GPT-5.4 finished 99.4% and met the task accuracy criterion in 47.0%. In ShoppingBench, GPT-4.1 met strict criteria in 59.6% of 250 product finder tasks and 30.4% of 250 coupon and budget tasks. Verify the exact variant and the condition the shopper needs before sending a link. Task completion versus accuracy In the 2026 EComAgentBench v3 study of 662 valid shopping tasks, GPT-5.4 finished 99.4% but scored 47.0% on task accuracy, defined as exact target match or satisfaction of every rubric. The measures share the benchmark task set but describe different outcomes. https://arxiv.org/html/2606.17698v3 Constraints from the opening and later turns On EComAgentBench v3, GPT-5.4 satisfied 88.1% of rubric checks sourced from the opening query, 69.8% sourced from the shopper profile and 70.9% from clarification, using the 2026 benchmark’s scripted shopping scenarios. https://arxiv.org/html/2606.17698v3 Finding a product versus satisfying an offer ShoppingBench v1 (2025) reports 59.6% strict success for GPT-4.1 on 250 product finder tests and 30.4% on 250 coupon and budget tests. The sandbox used real product records and different success conditions for the two intents. https://arxiv.org/html/2508.04266v1 Average relevance and complete success For GPT-4.1 on ShoppingBench v1’s 250 product finder tests, the reported average product relevance was 83.6% and strict success was 59.6%. The former credits matching attributes while the latter requires a full relevance score. https://arxiv.org/html/2508.04266v1 A recommendation can link to the wrong destination AgenticShop’s 2026 product-attribution analysis reports failure rates of 24.76% for ChatGPT Search and 30.00% for Claude Sonnet 4 Search. The paper distinguishes invalid pages from irrelevant destinations in its benchmark. https://arxiv.org/html/2602.12315v1 An evaluator needs product evidence as well as a URL In AgenticShop’s 50-instance human comparison, Pearson correlation with the human ratings was 0.7320 for the full evidence approach and 0.3792 for URL-only evaluation. Human-to-human correlation was 0.7736. Correlation measures agreement in scores. https://arxiv.org/html/2602.12315v1 Screenshots alone did not recover the same agreement In AgenticShop Table 3, the grading approach with a URL and screenshots but no extracted product information correlated 0.4395 with human scores, compared with 0.7320 for the full approach over the 50-instance study. https://arxiv.org/html/2602.12315v1 How much the ranking depended on what the grader could see In AgenticShop’s 50-instance evaluator study, the full evidence approach had Spearman correlation 0.7425 with human grades, versus 0.3885 for the URL-only approach. This is rank agreement on benchmark examples. https://arxiv.org/html/2602.12315v1 Shopping scores changed when the title was removed In the displayed ShoppingBench cells, explicit-title scores exceeded attribute-specification scores by 2.46 points for ChatGPT Search and 1.49 for Claude Sonnet 4 Search. These differences compare benchmark task conditions. https://arxiv.org/html/2602.12315v1#S5.T4 GPT-5.4 and Opus traded leads across sources of shopper intent In EComAgentBench Table 4, GPT-5.4 led Opus 4.6 by 7.5 points on query-visible constraints, while Opus led by 3.0 points on persona constraints. GPT-5.4 led by 2.3 points on clarification constraints. https://arxiv.org/html/2606.17698v3 The visible-to-persona gap differed by 10.5 points The query-visible minus persona rubric gap was 18.3 percentage points for GPT-5.4 and 7.8 for Opus 4.6 in EComAgentBench. The difference between the two gaps was 10.5 points. https://arxiv.org/html/2606.17698v3