Sort candidate sources
Run your source list through Gate. Sources that refuse drop out at once.
Building a dataset to sell or supply to others is one of the five purposes the RightSignal Algorithm decides, and it is the one treated with the most care. Supplying content to others carries it further from the site than any other use, and few sites address it in terms.
For this purpose RightSignal does not let silence stand as an answer.
This shapes what RightSignal offers a dataset company. It does not hand you a list of open sources. It separates the sources that have refused from the ones that need a person's judgement, and it gives that person the evidence to judge with.
Run your source list through Gate. Sources that refuse drop out at once.
For each held source, the decision quotes what the site says and names what could not be read. Your reviewer, with your lawyer where needed, decides on a record and not on a guess.
Store the receipt number with each item, so every row of the dataset links to the decision behind it.
Your buyer receives the data and the means to verify where it came from.
Click the scene for the next step
A buyer that receives receipts can verify them with the standalone verifier, with no connection to RightSignal and no need to trust you. A buyer that is a model provider serving the EU can describe, in its public summary of training content, what its supplier did to respect reservations. Your evidence becomes part of your buyer's answer, which is a reason to buy from you.
In early access
The ready-made policies in early access cover AI training and text and data mining. A policy for dataset building is being added. Tell us in your access request that you need it.