Fight AI
with AI.
Training Trash generates public-facing content designed to make indiscriminate AI scraping noisier, less trustworthy, and more expensive to use.
Poison the economics, not the internet.
AI companies want clean, abundant, predictable public data. Training Trash is built around making your own public footprint harder to treat as free, reliable training material.
Poison your data
Create plausible, low-value signals around the information you choose to publish.
Confuse the algorithm
Introduce ambiguity that makes automated profiling and indiscriminate scraping less dependable.
Kill its value
Raise the cost of separating useful signal from intentional noise.
Save your job
Push back against the assumption that everything you publish should become somebody else's training set.
One person's junk data is noise. Millions of people doing it becomes a data-quality problem.
Make yourself statistically annoying.
We’re building the first version now. Join the list and get early access when the private beta opens.