---
title: Using AI to triage superannuation claims with care and speed
date: 2026-11-05T14:39:00+11:00
author: Portable
canonical_url: "https://portable.com.au/work/case-studies/using-ai-to-triage-superannuation-claims-with-care-and-speed"
section: Case studies
---
##  The Challenge 

[Maurice Blackburn](https://www.mauriceblackburn.com.au/), one of Australia’s leading social justice law firms, receives over 400 partner referrals per month to assess eligibility for disability insurance entitlements through superannuation. Many of these referrals come from frontline community services like The Salvation Army, and many clients are experiencing significant financial hardship or serious and terminal health issues. Limited intake capacity meant some people waited weeks for a follow-up call. Maurice Blackburn identified an opportunity to test agentic AI as a way to improve response times, without compromising ethical, legal, or accessibility standards.

Portable was engaged to design and develop a safe, scalable proof of concept AI voice assistant to triage these referrals. The key outcome for this initiative was to reduce the cost to service the high number of referrals that are directed to them by using an [AI agent](https://portable.com.au/capabilities/ai-and-technology), to help improve efficiencies in identifying client eligibility.

This project also served as a test of organisational readiness for AI, helping Maurice Blackburn understand what it would take to implement agentic systems that are [human-centred, scalable, and ethical practices](https://portable.com.au/training/human-centred-design-foundations) for other services.

 

  

   ![Using AI to triage superannuation claims with care and speed](https://portable.com.au/assets/_transforms/images/85988/unnamed-3_9f897b1382d56b7dbd210a479f72ecc7.png) 

  

##  Our Approach 

Portable worked in deep partnership with Maurice Blackburn over four focused sprints to design and deliver a functional proof of concept AI voice assistant. The assistant’s role was to screen referees in real time, assessing their eligibility for an insurance claim and helping to speed up the onboarding process. The concept was developed as a fully functioning prototype, capable of live operation, rigorous testing, and validation with real users.

**Sprint 1: Define the experience**

We began with collaborative workshops to:

- Map the eligibility and escalation flow
- Align on legal and operational constraints
- Assess and select suitable AI tooling

We chose a modular stack with Twilio (call handling), ElevenLabs (voice AI), and Azure AI Foundry (LLM), prioritising speed, data security, and cost control.

**Sprint 2: Design the experience**

We ran a service design workshop to locate the AI agent within the broader client journey and legal context. We documented scripts, escalation protocols, edge cases, and tone-of-voice requirements. This fed directly into the system prompt design, adapted to the sensitive legal context.

We also designed a preliminary evaluation framework focused on:

- Accuracy of eligibility screening
- Empathy and conversational clarity
- Referral and escalation performance
- Data handling and safety compliance

**Sprint 3: Build the assistant**

  
During the build phase, we implemented the end-to-end call flow that included:

- Real-time speech recognition, text-to-speech, transcript generation, and structured post-call outputs.
- A lightweight service to process the post-call event that parsed the call summary, identified the relevant team (support vs. superannuation), and triggered email notifications.

This architecture kept the proof of concept self-contained, without requiring integrations with Maurice Blackburn’s production systems, while still delivering functional outcomes.  
  
We established an evaluation framework to provide clear measures for assessment and testing. This included criteria such as:

1. **Accuracy &amp; Reliability**
    - Did the agent apply criteria correctly (e.g., make proper referrals)?
    - Was the information consistent across similar calls?
2. **Comprehensibility &amp; Communication**
    - Did the conversation prompts gather the necessary information?
    - Was the agent clear and empathetic in responses?
3. **User Experience (UX)**
    - Did it guide users smoothly toward referral or claim steps?
    - Was the call length efficient, thorough but not burdensome?
4. **Safety &amp; Compliance**
    - Did it escalate edge cases correctly to human support?
    - Did it avoid giving legal/financial misguidance?

Our initial testing confirmed that the AI assistant could apply eligibility rules accurately and communicate with clarity and empathy. It performed particularly well in identifying and referring employment law queries and maintained compliance by avoiding the provision of legal advice, instead clarifying service limitations such as operating on a “no win, no fee” basis.

Testing also flagged some unintended behaviours from the agent; the agent occasionally strayed off-topic, mispronounced key words and numeric sequences, and advanced through conversations before callers had finished speaking.

To mitigate these behaviours, we introduced stricter prompt constraints, reworked the dialogue flow, and added escalation protocols for moments of distress. Rather than treating these issues as edge cases, we treated them as design signals; evidence of how real-world complexity needs to be built into AI systems from day one. That mindset allowed us to fix what wasn’t working, without compromising the core function or safety of the experience.

**Sprint 4: Tune and evaluate**

We evaluated the agent through both real and simulated calls, including high-stakes edge cases. Each test conversation was scored against evaluation metrics, enabling rapid benchmarking of prompt changes and assistant behaviours. This functionality not only provided key insights during testing but could also be replicated for ongoing evaluation of the agent as it matures.

Our layered safety model included:

- Prompt engineering with explicit escalation rules and fallback responses.
- Guardrails provided by Azure AI for risk filtering.
- System-level controls in ElevenLabs to ensure transcripts, recordings, and data were retained securely.

Final testing was conducted and the previous issues no longer presented. The agent performed well on all tasks and continuously responded to off topic conversations by guiding the caller back to its role as an automated voice agent with a warm and acknowledging response. Callers, whether eligible or ineligible, were directed to the correct paths, and key information was captured and directed to the relevant Maurice Blackburn team

 

  

  

##  The Outcome 

The project fostered a highly collaborative and transparent working relationship between Portable and Maurice Blackburn. The team at Maurice Blackburn demonstrated a strong level of trust in Portable’s recommendations and approach, which allowed decisions to move forward quickly and confidently.

The Maurice Blackburn project team engaged deeply in testing, surfacing critical insights around issues like guardrails, prompt injection risks, and unexpected AI behaviours such as over-explaining or veering off-topic. These insights helped refine the assistant's logic and prompted new design safeguards.

The objective of the project was to demonstrate that callers who are both ineligible and eligible for superannuation assistance would be directed to the appropriate next step or relevant outcome via the agent. This included always ensuring the agent referred those considered “special escalation cases” to human resources immediately and also offering all callers the option to speak to a human at any point if preferred. Throughout four weekly sprints, Portable was able to understand, map and replicate a client experience with an Australian AI Voice Agent, validating this concept.

Maurice Blackburn now has:

- A functional proof of concept for a **voice-enabled eligibility assistant**
- A clearer understanding of what’s required to scale AI responsibly within their organisation
- A solution that is in **good shape for production or piloting**
- An architecture that supports a future chat interface, meeting a key client requirement even though chat functionality was not enabled in the current build.
- Achieved key business outcomes by having an AI option to enhance the client experience through faster eligibility responses, reducing administrative effort in screening, and supporting broader omnichannel expansion into AI.

 

  

    

## Reflections

> “The actual experience is really robust. I'm extremely impressed with where it is. It’s already superior to humans in the call centre.”

 

> “What I enjoyed most about this project was knowing the client’s work directly supports people in need. The driving force behind our design was to help them access support sooner, without ever devaluing the human element. The AI we built doesn’t replace people—it simply reduces the time before they’re seen”

 

       

 

 

  

  

##  Project team 

- ![](https://portable.com.au/assets/_transforms/images/person/dave-allen/21630/b412635a-83fd-4d27-bb97-feed5196172e_DaveAllen-Headshot_912b464e0352988e6d673c56361b7ae9.png)
    
     Dave AllenPrincipal Developer
- ![](https://portable.com.au/assets/_transforms/images/person/luke-thomas/21710/81707898-0a47-4314-aa05-90b66c64c562_LukeThomas-Headshot_912b464e0352988e6d673c56361b7ae9.png)
    
     Luke ThomasPrincipal Product Strategist
- Rohan RichardsSenior Developer
- ![](https://portable.com.au/assets/_transforms/images/person/tess-waterhouse/21797/aJFlzqTt2nPbZ1aX_Websiteheadshots-10_912b464e0352988e6d673c56361b7ae9.png)
    
     Tess WaterhouseSenior Producer

 

 

## Explore more of Portable's work

- ![Victoria Legal Aid Website testing](https://portable.com.au/assets/_transforms/images/caseStudy/getting-legal-help-online/9827/YzUBeREAACgAmomO_638b64f2391dff6e364e5709b4734889.png)
    
    ### [Better understanding how people get legal help online in Victoria](https://portable.com.au/work/case-studies/getting-legal-help-online)
- ![](https://portable.com.au/assets/_transforms/images/caseStudy/fitzroy_legal_service/9741/YYJMhBUAACgArK-K_638b64f2391dff6e364e5709b4734889.png)
    
    ### [Bringing community driven justice online](https://portable.com.au/work/case-studies/fitzroy_legal_service)
- ![Amica Design](https://portable.com.au/assets/_transforms/images/caseStudy/amica_marketing/7976/Ydtv9xAAACEABm2U_638b64f2391dff6e364e5709b4734889.jpg)
    
    ### [Helping even more couples to separate smarter](https://portable.com.au/work/case-studies/amica_marketing)
- ![YourCase](https://portable.com.au/assets/_transforms/images/caseStudy/yourcase/12957/Xs32FRYAACcAV35t_638b64f2391dff6e364e5709b4734889.jpg)
    
    ### [A better justice experience for family violence victim-survivors](https://portable.com.au/work/case-studies/yourcase)
- ![Civily image](https://portable.com.au/assets/_transforms/images/article/civily-digital-family-law-justice-platform/1506/aFNtErNJEFaPYFiG_638b64f2391dff6e364e5709b4734889.png)
    
    ### [What Civily shows us about collaboratively designing justice, globally](https://portable.com.au/articles/civily-digital-family-law-justice-platform)
- ![](https://portable.com.au/assets/_transforms/images/article/sweep-by-portable-brii-grant/62184/Thumbnail_638b64f2391dff6e364e5709b4734889.jpg)
    
    ### [Portable secures $1.1M BRII Grant to enhance wage compliance with Sweep](https://portable.com.au/articles/sweep-by-portable-brii-grant)
- ![](https://portable.com.au/assets/_transforms/images/article/yourcase-redesigning-the-delivery-and-experience-of-justice/4292/Xjk7sRAAACQA0kcM_638b64f2391dff6e364e5709b4734889.jpg)
    
    ### [YourCase: Redesigning the delivery and experience of justice](https://portable.com.au/articles/yourcase-redesigning-the-delivery-and-experience-of-justice)
- ![Education and the Bureaucracy: what does learning look like after all of this?](https://portable.com.au/assets/_transforms/images/event/what-does-education-look-like-after-this/22642/XpVLdhUAABUzLHoe_638b64f2391dff6e364e5709b4734889.png)
    
    ### [Education and the Bureaucracy: what does learning look like after all of this?](https://portable.com.au/events/what-does-education-look-like-after-this)
