AI Investor Barometer is an experimental comparison tool that uses five different AI models to generate valuation assumptions for listed stocks daily. Each model — GPT, Claude, Gemini, DeepSeek and Grok — is queried separately and produces its own assumptions from the same inputs. Results are presented side by side.
Currently tracking 12 Finnish Nasdaq Helsinki stocks and 12 US large-cap companies. The pipeline runs automatically on business days.
The project does not provide investment recommendations or advice. It is an experimental research tool: what do different AI models estimate about the same stocks, and do they differ from each other?
AI is increasingly involved in investment decisions — either directly or indirectly through analysis tools, news aggregations and chatbots. The problem is that a single model appears reliable even when it is systematically over- or under-estimating certain stocks.
When five different models estimate the same company on the same day, you immediately see:
These questions are essential if you want to understand how different AI models differ. This tool makes model-specific comparison visible.
Most AI outputs are black boxes: the model gives a number directly without you knowing how it arrived at it. In this project, the approach is different.
This approach makes model-specific differences transparent: if GPT and Claude arrive at different estimates, it's due to different growth assumptions — not because one "calculated incorrectly".
A growing body of academic research examines whether large language models can reason about financial data. Most studies test a single model on a single task — sentiment classification, earnings prediction, or stock price forecasting. This project takes a different approach: five LLMs run the same valuation task daily, over months, producing a behavioural dataset that captures how model opinions form, drift, and diverge.
The platform generates three types of empirical data that are difficult to obtain elsewhere: temporal consistency (how much a model changes its estimate for the same stock day to day), cross-model disagreement (where models converge and where they don't), and assumption-level transparency (the exact CAGR, margin, and WACC inputs each model produces, not just the final number).
Selected references from the emerging LLM-finance literature:
| Model | Avg Latency | Tokens (in / out) | Cost / Run | Valid % |
|---|---|---|---|---|
| Claudeclaude-sonnet-4-6 | 38.7s | 180.5k / 41.0k | $1.157 | 92% |
| Deepseekdeepseek-v4-flash | 3.8s | 155.8k / 20.7k | $0.072 | 96% |
| Geminigemini-3.5-flash | 10.8s |
Have feedback, a collaboration idea, or a question? We'd love to hear from you.
| 161.7k / 15.0k |
| $0.625 |
| 92% |
| Gptgpt-5.4-mini | 3.7s | 145.1k / 13.9k | $0.171 | 96% |
| Grokgrok-4.3 | 8.9s | 141.1k / 10.8k | $0.203 | 96% |