People
Policy documents that stay current and show their changes
A policy nobody has updated in three years is worse than no policy. A worker keeps the drafts moving and the versions visible. A qualified adviser signs them off.
What the worker does
An AI worker in Polaris drafts and maintains HR policy documentation in versioned Polaris docs: writing a first draft from your notes, marking what changed between versions, collecting comments from reviewers, and flagging policies that have not been reviewed within the period you set. Whether a policy is lawful where you operate is a question for a qualified adviser, and Polaris does not answer it.
- Lives in
- Versioned Polaris docs
- Connections
- Google Drive · Slack · web search
- Sign-off
- A named human, every time
The failure mode is silence, not error
Most handbooks are not wrong on the day they are written. They go wrong by sitting still: the company moves to four countries, adopts a different working pattern, changes who approves expenses, and the document describes a company that no longer exists.
Nobody updates it because updating it means finding the current version, working out what changed, writing the change, getting someone to approve it, and telling everyone. Five steps, no owner.
A worker can carry four of those five. The approval stays with a person, and it should be a person who is qualified to give it.
What the worker maintains
The draft
A first version written from your bullet points and your existing document, in your own house language rather than boilerplate lifted from somewhere else.
The diff
What changed since the last version, stated in a sentence per change, so a reviewer reads the delta and not the whole document.
The review clock
Any policy past its review date appears as a dated task in the Focus lane with an owner attached.
The comment round
Reviewers comment on the doc; the worker collects unresolved comments into a list and chases the people who have not responded.
The publication note
A drafted announcement of what changed and what people need to do differently, for a human to send.
Draft versus decide
The worker prepares
- Wording, structure and consistency across documents
- Cross-references between related policies
- A summary of what an existing policy currently says
- Research on the open web, with every source named
- Version history and change lists
A qualified person decides
- Whether the policy is lawful in each place you employ people
- What the company's position actually is
- Which obligations you are accepting
- When the policy takes effect
- Approving and publishing it
Questions people ask
+Can the worker tell us whether our policy is compliant?
No. It can tell you what your document says, what it does not cover, and what public sources it found on a topic, with each source named. Whether that meets your obligations in a specific jurisdiction is a judgment for a qualified adviser.
+Why keep policies in Polaris rather than a wiki?
Because the review cycle is a task, and the task and the document then live in the same place. Docs are a nested tree with versioning and file review, and a policy past its review date can be a dated item in someone's Focus lane instead of a note in a spreadsheet.
+How do we stop the worker inventing a policy position?
Brief it to draft only from material you supply, and to write an explicit open question wherever your notes do not settle a point. A visible list of open questions at the top of a draft is far more useful than a confident paragraph that quietly guessed.
+Can it write the announcement too?
Yes, as a draft. Anything that goes to the whole company about how people are employed is sent by a person, from their own account, after they have read it.
Related
HR work an AI worker can prepare, and where it must stop
Scheduling, onboarding logistics, policy drafts and review-cycle admin, prepared by a worker. Every decision about a person stays with a person.
Onboarding that is finished before the first Monday
Nobody's first day should start with an apology about accounts. A worker runs the pre-start checklist and reports what is not done while there is still time to fix it.
Knowing that a source changed, on the week it changed
A worker watches the public pages you nominate, quotes what changed, and dates it. What the change means for you is a question for your adviser.
Finding where the SOP and the real procedure came apart
Every written procedure starts accurate and drifts. A worker compares the document to how the work actually ran and reports the difference.
Hire an AI technical writer
The documentation debt on your team is not a writing problem, it is a nobody-has-two-free-hours problem, and this is the worker for those two hours.
Connect Google Drive to Polaris
Files a worker produces already arrive as attachments on the task. Files your team already keeps in Drive are the part still waiting.
Web search in Polaris
Nothing to authorize, nothing to store, and every query a worker runs appears in the activity feed while it works.
Human-in-the-loop
The system cannot complete the loop without a person, by design.