Overview

Versioning, history, and provenance are first-order concepts within the HASH database, which provides:

1. →

   a complete record of all the changes to an entity or type over time (change tracking)

2. →

   the ability to inspect, access and even use older versions of an entity or type (versioning)

3. →

   the ability to see where information and changes came from, as well as what caused them (provenance)

In a world of increasingly AI-generated information, the ability to map the "supply chain" of your information and beliefs is critical, and amongst open-source tools only HASH is well-placed to help in this regard.

Using provenance metadata

AI usage

HASH AI, including HashGPT, uses provenance data to return answers to you that are accurately caveated, and which can be introspected further. Rather than blindly tell you that things are or are not true with a false sense of certainty, provenance metadata is incorporated into answers to inform confidence-based judgements, and give answers to thorny, real-world issues that can be relied upon similar to a human analyst or co-worker's judgement, with a realistic and appropriately conveyed sense of doubt, where the information required to arrive at a judgement is unclear or incomplete, or its provenance is less than certain.

Direct usage

In addition to AI being able to access and factor in provenance metadata, you can also view and access it directly to use it as part of your analysis. Simply head to the "Context" tab of a given entity or type to view its change history and any other available provenance metadata. This metadata is collected and made accessible web-wide.