What AI Knows About You
Discover what AI models know about you based on their internal knowledge.
Last updated
February 10, 2026
Connectors used
Tags
Key Takeaways
- 3 AI models queried in parallel - GPT, Gemini, and Claude each respond independently
- No web search - Only queries what models already know from their training data
- No private data accessed - Does not scrape websites, social media, or private databases
- Try your own name - Enter any name and compare what each model returns
- Customizable - Modify the prompt or add more AI models to the workflow
What This Workflow Does
This workflow sends the same prompt to three AI models (GPT, Gemini, Claude) in parallel and returns their responses. You enter a name, and each model independently shares what it knows based on its training data.
Use cases:
- Check if you appear in AI training data
- See what AI models know about public figures
- Compare how different models respond to the same person
- Explore differences in AI training datasets
How It Works
| Step | What Happens |
|---|---|
| 1. Manual trigger | You enter a name |
| 2. Parallel AI queries | Same prompt sent to GPT, Gemini, and Claude |
| 3. Compare results | View each model's response |
Each model responds independently without seeing the others' answers.
AI Models Used
| Model | Provider |
|---|---|
| GPT | OpenAI |
| Gemini Flash | |
| Claude Sonnet | Anthropic |
Setup Instructions
- Click "Use template" on this page
- Open the manual trigger node
- Enter a name
- Run the workflow
- Compare the three responses
Tips for Better Results
- Start with your own name and see what comes back
- Common names may return results about someone else. Add context like your profession or city to narrow it down
- Public figures will generally return more detailed results than private individuals
- Try different prompts by editing the AI nodes to ask more specific questions
When Models Disagree
Different responses across models can indicate:
| Reason | Explanation |
|---|---|
| Training data differences | Each provider uses different data sources |
| Cutoff dates | Models have different knowledge cutoff dates |
| Ambiguous names | Models may match different people with the same name |
| Uncertainty handling | Models differ in how they handle information they are unsure about |
Customization
| What You Can Change | How |
|---|---|
| Prompt | Edit the prompt in any AI node to ask different questions |
| Models | Duplicate an AI node and select a different model |
| Output format | Change the structured output schema in the AI node settings |
Privacy
| What this workflow does | What this workflow does not do |
|---|---|
| Queries AI models' existing knowledge | Search the web |
| Returns what is already in training data | Access private databases |
| Runs in your Needle workspace | Scrape websites or social media profiles |
FAQ
Q: Will I appear in AI training data? A: Only if you have a significant public presence online. Most private individuals won't appear.
Q: Can I remove myself from AI training data? A: Not from existing models. Some providers offer opt-out for future training.
Q: Why do models give different answers? A: Training datasets, cutoff dates, and handling of uncertainty differ across models.
Want to showcase your own workflows?
Become a Needle workflow partner and turn your expertise into recurring revenue.
