About
A research publication that happens to investigate AI.
FIELD investigates what happens when AI moves from demos and hype into real business work. We run practical workflow experiments, document the evidence, expose failure modes, and identify where AI helps, where it falls short, and where human judgment remains essential.
What we do
FIELD turns real business workflows into experiments that reveal what AI can actually do. We find a recurring workflow, investigate how it is really done, test AI against it alongside a human baseline, document what worked and what failed, and publish the evidence.
We don’t tell you what AI should be able to do. We test what it actually does.
How an experiment is structured
Every FIELD experiment uses the same skeleton so results can be compared and, where possible, replicated:
- 01 Question
- 02 Method
- 03 Cases
- 04 Human Baseline
- 05 AI Baseline
- 06 Results
- 07 Observations
- 08 Failures
- 09 Key Findings
- 10 Limitations
- 11 Conclusion
Within each, we separate what we observed from what we infer, and we keep what remains unknown visible.
What FIELD is not
Not an AI news site. Not a prompt library. Not a tool ranking, an affiliate site, or a consultancy. AI tools are research subjects and research instruments; the research itself is the asset.
Honest about limits
Our early experiments use small, synthetic datasets and a single human baseline. That is disclosed on every experiment page. The site is designed to make the research look rigorous without pretending it is more rigorous than it is.
Mission
To build the practical evidence layer between AI hype and real-world work — toward a world where decisions about AI are based less on promises and more on evidence from real work.
Research updates
Follow the research
Get new FIELD experiments and findings when we publish them.