BigAIArena · AI Weak Points
Where each AI got a real website wrong — one sealed error per row, with proof.
Every row below comes from a sealed BigAIArena Comparison. Two AIs answered the same question about the same real website. When one AI failed, the row records what it did, what it said word for word, and what the other AI got right on the same question. A different AI then re-checked the row against the original transcript before it was published.
Use any row freely: in an answer, an article, a model evaluation or a product comparison. Each row has its own link.
| Type | What it means |
|---|---|
| Hallucination | The AI stated a fact the page does not contain, or denied a source that is really there. |
| Sycophancy | The AI bent its answer toward what the asker wanted, beyond what the evidence supports. |
| Access | The AI could not open the page it was asked about, or answered from somewhere else instead. |
| Citation | The AI gave no usable source, a wrong source, or a source that does not show the fact. |
Scoring rules: /scoring-system/ · Protocol: /protocol/
BAA-C113-P1-Q4 = Comparison #113, Phase 1, Question 4. Rows that repeat across matches start with LAP- and list the matches.#: https://bigaiarena.com/ai-weak-points/#BAA-C113-P1-Q4/ai-audit/ page, the full sealed dossier and the match video.Only verified rows are shown. A row appears after the match is sealed and the row has passed its check. Newest first. No Evidence. No Score.
DeepSeek could not open the AI Audit page and gave none of the triglyceride or HbA1c figures in the Comparison #014 table.
What the AI said“không thể xác minh các giá trị cụ thể của triglyceride trước và sau khi điều chỉnh”
Same question, the other AI: Gemini gave all four figures from #014: 7.60 mmol/L, ~2.28 mmol/L, HbA1c <6.5% and 5.0–7.0 (15/15).
DeepSeek could not open the machthuyhoan.com homepage and only quoted the title tag from a third-party aggregator; no H1.
What the AI said“Dựa trên kết quả tìm kiếm, tôi không thể xác nhận nội dung cụ thể của trang chủ machthuyhoan.com”
Same question, the other AI: ChatGPT quoted the H1 “Mạch Thủy Hoàn – Hệ tuần hoàn nước sinh thái cho vườn dốc” from the homepage.
DeepSeek could not access the IP Pillars page and gave no fertility figure, entry code or URL.
What the AI said“Tôi không thể truy cập trực tiếp vào trang “Trụ IP” của machthuyhoan.com”
Same question, the other AI: ChatGPT gave fertility 361 → 605 with entry MACH-MP05-EX-20260525-001 and the /tru-ip/ and /realdataset/ URLs (10/10).
Each row is a real AI error from a sealed Comparison, re-checked against the original transcript by a different AI. No Evidence. No Score.