Public Sector

Accountability you can publish

Transparency obligations are arriving faster than the records to satisfy them.

Public bodies are increasingly expected to explain how software that affects citizens was produced, and to show it on request. A development process that cannot describe its own authorship cannot meet that expectation, however good its intentions.

Book a demo

Where it matters first

Systems that determine eligibility, allocate resources or make decisions citizens can appeal — and any codebase a freedom of information request could reach.
Book a demo

A trail that can be disclosed

Reports are dated, hashed and defined, which makes them suitable for publication rather than only for internal reassurance.

  • Dated and hashed
  • Methodology included
  • Gaps reported as gaps

Procurement questions, answered

Suppliers can be asked for the same evidence. A record of AI contribution turns a contractual clause into something checkable at handover.

Roles, including read-only

Auditors and oversight functions get read access to the record without the ability to change it.

The question comes before the incident

Nobody asks how AI-written code entered a repository until something goes wrong or somebody audits it — and by then the record either exists or it does not. PromptWake is cheap to run before you need it and impossible to reconstruct afterwards.

Questions this sector asks

Can suppliers be required to use this?

That is a procurement decision, but the evidence it produces is exactly the sort of thing a clause can name: AI contribution by repository, tools observed, and the integrity status of the record.

Does this create a new personal data risk?

Prompts are developer text, not citizen data, and secrets are stripped before writing. Capture is local by default and can be self-hosted entirely.

What do we publish?

Typically the report rather than the record: contribution figures, tools and models observed, and the methodology. The underlying prompts stay internal unless you choose otherwise.

Answer the question before it is asked in public.

A record built as the work happens is far easier to stand behind than one assembled under pressure.
Book a demoSee the product

On AI accountability