Sentiment tracking across AI answer engines
Sentiment in AI answers is how an engine characterises a brand when it names it. Appearing in an answer as the cautionary example is not the same as being the recommendation, and citation counts cannot tell the two apart.
How it differs by engine
Derived from the answer text with the classifying sentence retained as evidence.
Google AI Overviews sentiment
Derived from the summary text, with the classifying sentence retained.
google_ai_overviews
Google AI Mode sentiment
Derived from the answer text with the evidence sentence retained.
google_ai_mode
ChatGPT sentiment
Derived from the answer text with the classifying sentence retained.
chatgpt
Perplexity sentiment
Derived from the answer text, with the evidence sentence and its citation retained.
perplexity
Microsoft Copilot sentiment
Derived from the answer text with the evidence sentence retained.
copilot
Gemini app sentiment
Derived from the answer text with the supporting sentence retained.
gemini
DeepSeek sentiment
Derived from the source-language answer text with the evidence sentence retained.
deepseek
Qwen sentiment
Derived from source-language answer text with the evidence sentence retained.
qwen
Doubao sentiment
Derived from the source-language answer text with evidence retained.
doubao
Kimi sentiment
Derived from the source-language answer text with the evidence passage retained.
kimi
Baidu ERNIE sentiment
Derived from source-language answer text with the evidence sentence retained.
baidu_ernie
Naver sentiment
Derived from the source-language answer text with evidence retained.
naver
Sampling
What the API returns
{
"engine": "google_ai_overviews",
"field": "sentiment",
"prompt": "best ai visibility tracking tool",
"country": "GB",
"language": "en",
"n": 30,
"sentiment": [
{
"brand": "Nostimates",
"label": "recommended",
"confidence": 0.86,
"distribution": {
"recommended": 0.6,
"neutral": 0.33,
"hedged": 0.07
}
}
]
}Response fields
| Key | Type | Notes |
|---|---|---|
| brand | string | Brand the classification applies to. |
| label | string | recommended | neutral | hedged | negative. |
| confidence | float | Classifier confidence for this sample. |
| evidence | string | The sentence the label was drawn from. |
| distribution | object | Label shares across the n samples. |
Sentiment FAQ
- What is sentiment tracking?
- Sentiment in AI answers is how an engine characterises a brand when it names it. Appearing in an answer as the cautionary example is not the same as being the recommendation, and citation counts cannot tell the two apart.
- How is sentiment sampled?
- Classifications are reported as distributions across n, because the same engine will frame the same brand differently between samples.
- What does sentiment cost?
- Add-on field. Priced per collection at a small multiple of the base credit.
Get sentiment on your prompt set
Track it in the dashboard or pull it in through the API. Send us the prompts, the brands and the markets, and we'll come back with a per-engine coverage plan and the credit arithmetic behind it.