MUXA MUXA Design Studio
Case studies Authorium Case Study

Designing a compliant AI editor for government workflows

An AI-powered document editor for government teams that brings efficiency and transparency to budgeting, procurement, and grant management, without compromising compliance.

Client
Authorium
Sector
Government technology
Role
UX UI, User Research, Product Design
Focus
AI document editing, Compliance, Collaboration
Year
2024

Streamlining complex government workflows

Authorium helps government agencies manage complex processes such as budgeting, procurement, and grant management, making their work more efficient, compliant, and collaborative.

Much of that work runs on long, formal documents. A request for proposal, or RFP, is the document an agency publishes to invite vendors to bid on a contract; grant management and procurement generate stacks of similar paperwork. Each document passes through many hands and has to satisfy strict legal and regulatory rules before it can be published.

Authorium aimed to create a secure, AI-powered document editor for that environment. The challenge was specific: integrate generative AI, software that can draft and rewrite text on its own, without compromising accuracy, confidentiality, or regulatory compliance. In a regulated document, an AI suggestion that is only "almost right" is not a small error; a single unverified change can fail a compliance audit.

We partnered with Authorium to design a solution that balanced innovation and accountability. Trust is not a finishing touch here, it is the prerequisite for adoption, so the work centered on transparency and control in every interaction: clear attribution on AI contributions, audit trails that survive review, and notifications that separate compliance signals from everyday chatter.

Fig 01 The editor: a navigable table of contents on the left, the document in the center with AI-touched passages clearly highlighted, and collaborator comments on the right. Locked sections and saved-state are visible at a glance.
40%
Faster
Expected document creation time for RFPs and templates
+40%
User satisfaction
In collaborative editing workflows (prototype testing)
0
Compliance breaches
Anticipated through audit trail and access control
+50%
Trust
In AI content through clear labeling and citations

Figures reflect Authorium's internal prototype-testing benchmarks. Production rollout metrics will replace these once live.

Every task anchored to the document

The Tasks system helps teams stay organized and accountable when collaborating on complex documents. Each task is tied to highlighted text or a specific section of the document, never floating free, and can include AI-generated suggestions, manual edits, or review requests. When a reviewer can see exactly which passage a task refers to, they no longer have to guess where a change came from.

  • Assign tasks directly within the editor for collaboration
  • Filter by priority, status, or assigned user
  • Link each task to its document context for clarity
  • Automatically notify users when tasks are completed or require review
Fig 02 Tasks live on the sections they belong to. Status pills (In Progress, Locked) sit on each section heading, and comments on the right are tied to the exact passage under discussion.

Notifications without the noise

The Notifications panel centralizes all activity updates, from AI alerts to collaborator actions, creating a single source of truth for progress and compliance. The categorization is the point: a compliance alert can never be allowed to get buried under routine collaboration chatter.

  • Real-time alerts for edits, approvals, and AI-generated updates
  • Categorized notifications: AI suggestions, reviews, compliance alerts, mentions
  • Configurable preferences for the type and frequency of alerts
  • Inline contextual previews, so users can act immediately

This system reduces email clutter and keeps teams aligned, ensuring that no important update or compliance check is missed.

Fig 03 The notifications panel sorts events into streams, so AI suggestions, reviews, compliance alerts, and mentions each surface separately. Every item carries an inline preview and a one-tap action.

An audit trail that separates AI from human work

The Activity Log provides full transparency and accountability throughout the document lifecycle, which is crucial for regulated government environments. Auditors do not just need to know what changed; they need to know whether a human or a model made each change, and the log makes that distinction visible at the row level.

  • Chronological record of every change, comment, and approval
  • Distinguishes between human edits and AI-generated contributions
  • Shows who made each change and when
  • Allows auditors to export logs for compliance reporting

The Activity Log builds confidence in the system by ensuring every action is traceable, reviewable, and secure, which is a core requirement for government adoption of AI-assisted tools.

Fig 04 The activity log at the project level: a filterable, exportable record of every document action across the workflow stages, attributed to a named person and a timestamp.
In regulated AI work, traceability is not a bonus feature. It is the baseline that makes adoption possible.

Read another case study

More selected work across enterprise AI, complex internal systems, and consumer fintech.

PeakMetrics Atlassian Schedule a call