18th November 2026
5 min read
Practical AI, Not Sci-Fi
An updated perspective on practical AI in case management, exploring how client needs and real-world use cases are shaping future development.
Artificial intelligence has become one of the most talked-about topics in legal and public sector work. New tools appear weekly, headlines promise transformation, and there’s an increasing sense that organisations need to “do something” with AI... quickly.
At Iken, we think that instinct needs reframing.
The question isn’t how fast you adopt AI, or even how much of it you use. The real question is what purpose it serves, and whether it genuinely supports the people doing the work.
Start with the work, not the technology
Legal and public sector teams don’t struggle because they lack technology. They struggle because their time and attention are stretched across competing demands: case work, administration, operations, and reporting: all under pressure to deliver high-quality outcomes with limited resource.
When we look at case management, we tend to think in four interconnected areas:
Case work: the expertise-led work of case owners and supporting staff
Case administration: time recording, information capture, and maintaining case records
Case operations: building workflows, templates, and processes that support consistent delivery
Business operations: reporting, insight, and understanding performance at scale
AI is often positioned as a solution for case work, summarising documents, drafting content, or interrogating large volumes of text. In many respects, that area is already well served. There are excellent tools available that help professionals interact with complex information more quickly and confidently.
Our view is that case management software shouldn’t try to replace those tools. Instead, it should support and complement them.
"There’s a lot of pressure right now to ‘do something’ with AI in legal and public sector work. Our view at Iken is actually quite simple. AI shouldn’t be the objective. It’s one possible solution to a problem, and often not the only one."

Zak Chester
Chief Product Officer
AI is a solution, not the objective
At Iken, our philosophy around AI mirrors our broader product philosophy: support people in their roles, reduce unnecessary drudgery, and help teams focus on the value only they can bring; judgement, expertise, and perspective.
That’s why we’re cautious about positioning AI as an end goal. Something artificially intelligent is a possible solution to a problem, not the problem itself, and shouldn’t be the objective. If a challenge can be addressed more simply, more reliably, or more transparently without AI, then that’s probably the better approach. This is true of any new technology.
The purpose of AI in a case management context, in our view, and that of our clients, is narrow but powerful:
Making case administration easier and faster, so professionals spend less time on repetitive tasks
Simplifying case operations, so teams can focus on improving services rather than managing tools
This is where we see the most meaningful opportunity.
"Where we see the most value is in using tools, including AI, to reduce the admin and operational burden around case work."

Zak Chester
Chief Product Officer
Where case management and AI genuinely intersect
Case work (drafting, summarising, analysing documents) is already an active area for legal AI, and rightly so. Business operations, too, benefit from established reporting and business intelligence tools.
Where we believe case management systems like Iken can add the most value is in case administration and case operations.
This view has been validated through our direct work with clients, discussing where Iken and AI could reduce the demands of case administration and operations. The use cases raised in these conversations include:
AI time recording suggestion
When
Today 11.05 - 11.50
What
Document review
Why
Follow-up activity
45 min
Add to timesheet
Supporting more accurate and timely time recording
Using information from case interactions, Iken could provide AI with the context to determine when a time record should be created, what should be recorded, where it should be assigned and why that activity is pertinent to a case.
The aim is to make it as simple as possible to create a timesheet, improving time capture and accountability while reducing the administrative effort involved.
Potential Outcome
More accurate time capture with less manual input
AI analysis
Category: Access request
Priority: Standard
Missing: Preferred contact method
Human review
required
Speeding up case intake
A case request is going to come from somewhere: either from a client or through an automated or agentic process.
Rather than relying on a webform and automating the subsequent steps, a case description and supporting information could be provided for AI to review, create, document and initiate a case.
Different case types, with different initiation requirements, could be catered for. Gap analysis could help move cases forward without unnecessary review effort, while acknowledging the need for human involvement or oversight.
Adding a layer of workload review could also support more intelligent case assignment.
Potential Outcome
Faster case creation and smarter work allocation
AI summary
Summary: Recent activity has been reviewed and the case is ready for the next stage.
Creating clear, consistent case summaries
Case management systems are data-rich. They hold both granular and high-level information relating to an understood process.
With process and case context, AI could summarise the current case position and outline expected next steps, providing information that a downstream AI process might otherwise lack.
This could give case owners a clearer picture of progress and provide context for subsequent processes.
Potential Outcome
Clearer case visibility and quicker understanding of progress
Request revieved
Check information
Approval needed
Have the terms been agreed by all relevant parties?
Yes - Proceed
No - Revise
AI builder
Supporting workflow building and information capture
AI can help remove the 'blank page' when documenting a process, although organisational differences and requirements mean expert adjustment is likely to remain necessary.
If this capability could be expanded to create the in-system information capture and workflow, it could help teams deploy case management workflows to end users more rapidly.
The aim would be to realise the benefits of these systems with less effort and time, without removing the expertise needed to configure them appropriately.
Potential Outcome
Faster workflow configuration with less manual setup
From client conversations to product development
These are some of the opportunities raised in our conversations with clients. They are not necessarily features in development, but examples of where client requirements and potential technical solutions intersect. Those conversations help shape where we focus our product development.
This isn't about removing people from the process. It's about reducing the background effort required to keep work moving, so teams can focus on service delivery and improvement.
Experiment thoughtfully and govern properly
Experimentation with AI must always sit alongside strong governance. This is as true for public sector teams as it is private sector. Information security and privacy are non-negotiable. Any exploration of AI tools needs clear boundaries, clear ownership, and a shared understanding of risk.
One of the most effective approaches is to start with well-defined use cases. Form a hypothesis, test it carefully, and evaluate the outcome. Learning where AI doesn’t add value is also useful albeit without the added productivity benefit of where it does.
Sharing those findings internally across teams and roles helps build collective understanding and avoids fragmented adoption.
It’s also important to be honest about cost. AI has economic implications, from licensing to skills and oversight. It also doesn’t bring human judgement, contextual understanding, or lived experience, and it shouldn’t be treated as a substitute for them.
The wider data governance picture
Responsible AI development also means understanding the wider legal and regulatory environment, particularly where data access, sharing and portability are concerned.
The EU Data Act is one development in this landscape. Although it is not specifically an AI regulation, it introduces new requirements concerning certain data rights and switching between data processing services.
Iken's Commercial Contracts Lawyer, Helen Naylor, explores these changes in her forthcoming whitepaper, EU Data Act: What Do We Need to Know?, examining the practical considerations for organisations and technology providers.
Avoiding the wrong habits
Perhaps the biggest risk we see is expectation. AI won’t solve every operational challenge. It won’t fix unclear processes. And it won’t remove the need for professional oversight. Treating it as a universal answer could create more problems than it solves.
A more sustainable approach is simpler... understand the requirement, test carefully, measure the outcome, and decide what genuinely improves productivity.
That’s the mindset we bring to our own product development and operations, and the one we encourage our customers to adopt.
Practical AI isn’t about science fiction. It’s about using the right tools, in the right places, for the right reasons and always in service of the people doing the work.
"That’s what practical, purpose-led use of AI looks like to us, and it’s the approach we take in our own product development."

Zak Chester
Chief Product Officer





