The latest AI in Professional Services Community meeting brought together colleagues from across the sector to share their experiences of working with AI in their institutions. The discussion was wide-ranging, practical, and honest, focusing on the everyday realities of time pressures, governance barriers, skills development, and responsible experimentation.
Across the conversation, a clear message emerged: successful AI adoption in professional services is less about the technology itself but more about creating the right conditions for meaningful use – e.g. time, structure, safety, and collaboration.
The key discussion points were:
Finding time in a crowded working day
One of the strongest themes of the session was the challenge of finding time to learn and experiment with AI alongside already full workloads. Participants recognised that enthusiasm alone is not enough when calendars are stretched and priorities compete.
Rather than carving out entirely new time, many contributors recommended embedding AI exploration into existing work. This included experimenting with AI while completing routine tasks, attending community sessions as a form of structured learning, and using real problems, rather than hypothetical scenarios, to make learning immediately relevant.
Several participants reflected that shifting mindset was just as important as shifting schedules. Treating AI engagement as a legitimate part of professional practice, rather than an optional extra or personal interest, helped unlock progress and reduced the sense of “extra work” attached to learning.
Start with the problem, not the tool
A recurring insight was the value of a problem-first approach. Participants shared experiences of initially being drawn into “scattergun” exploration of multiple tools, only to find themselves overwhelmed.
The conversation highlighted the importance of clearly identifying the problem to be solved before selecting a tool. In practice, this meant starting with existing, licensed platforms (particularly enterprise tools already approved by institutions) and testing them thoroughly before seeking alternatives. This approach not only reduced risk but also helped teams focus their limited learning time more effectively.
Several contributors noted that structured reflection, often introduced through formal learning programmes, helped them recognise where they had previously jumped too quickly to solutions. Slowing down to articulate the problem created clearer use cases and more sustainable solutions.
Creating safe spaces to experiment
Participants strongly emphasised the need for safe environments for experimentation. With increasing awareness of data protection and security risks, many institutions are understandably cautious – but excessive caution can stall innovation.
The discussion highlighted practical ways to balance safety and experimentation. These included using synthetic or publicly available data, restricting access to early pilots, and clearly separating exploratory work from live operational systems. Creating designated “playgrounds” for AI testing allowed staff to build confidence while protecting sensitive information.
Safe spaces were also framed as cultural as well as technical. Contributors spoke about the importance of encouraging curiosity, allowing small failures, and normalising learning through experimentation rather than expecting immediate perfection.
Governance: Enabler or barrier?
Governance was one of the more complex topics of the session. Many participants expressed frustration at slow-moving institutional processes, particularly around data protection impact assessments (DPIAs), information security, and policy approval.
At the same time, examples were shared of governance structures that actively supported progress. Oversight groups with senior sponsorship were seen as particularly effective, helping to align innovation with accountability and providing visible leadership endorsement. Rather than creating entirely new AI policies, some institutions described success in updating existing policies (e.g. academic integrity, research governance, or data management) to include AI considerations.
A key insight was that governance works best when it’s iterative and collaborative, not static. Participants acknowledged that AI tools evolve too quickly for rigid frameworks.
Learning pathways that actually work
The session highlighted enthusiasm for structured learning routes, particularly apprenticeships and accredited programmes that integrate learning into paid working time. These approaches helped legitimise time spent on AI development and provided a clear framework for progression for staff.
Participants noted that such programmes offered more than technical skills. They encouraged reflective practice, better prompting techniques, and a stronger focus on evaluating value and impact. Importantly, structured learning also helped people avoid constantly chasing “shiny new tools” by grounding exploration in defined objectives.
Alongside formal routes, informal learning played a crucial role. Community sessions, peer discussions, short videos, and even everyday experimentation outside work were all cited as valuable ways to build confidence and understanding incrementally.
The power of community collaboration
One of the most optimistic threads running through the meeting was the emphasis on cross-institution collaboration. Participants highlighted the benefits of sharing principles, governance approaches, and lessons learned – particularly where institutions face similar challenges.
Examples were shared of regional and sector-wide groups working together to develop shared AI principles and frameworks. These collaborative efforts reduced duplication, accelerated progress, and provided reassurance that challenges were not being faced in isolation.
There was a strong interest in extending this collaboration further, including sharing anonymised templates, guidance, and governance artefacts where appropriate, to help the sector move forward collectively.
Data access, transparency, and trust
As AI tools become more deeply embedded in enterprise platforms, concerns around data visibility and access are becoming more pressing. Participants discussed how AI can surface information users may not realise they can access, raising questions about data labelling, permissions, and organisational readiness.
The conversation reinforced the importance of regular data audits, clear file labelling, and staff awareness of how AI tools interact with institutional systems. Rather than viewing this as an AI-specific issue, contributors noted that AI often exposes existing data management weaknesses that have long gone unaddressed.
Building trust, both with staff and leadership, was seen as critical. Transparency about the capabilities and limitations of AI, alongside practical guidance on responsible use, has proven to help reduce anxiety and improve ease of adoption amongst staff.
Keeping people engaged and curious
Engaging staff emerged as both a challenge and an opportunity. Participants shared creative approaches to sparking interest, including using everyday, relatable examples of AI use outside work and encouraging peer-to-per learning through buddy systems.
Several contributors observed that enthusiasm often came from unexpected places, challenging assumptions about who would engage most readily with AI. Small successes, shared informally, helped build momentum and confidence across teams.
Looking ahead
As the session drew to a close, participants reflected on the pace of change ahead. AI capabilities are evolving rapidly, and tools that feel limited today may become transformative within months. Documenting experiments, revisiting earlier use cases, and maintaining a learning mindset were all highlighted as essential strategies.
Above all, the discussion reinforced that AI adoption in professional services is not a single project or decision point. It is an ongoing journey – one that benefits from patience, shared learning, and a willingness to adapt.
Our next community session is on 11th February 3:30-4:30 pm.
Useful Links
Where AI fits: insights from Jisc’s internal AI impact workshop – Insights from the impact workshops run by the AI team here at Jisc to assess where AI can make an impact on the way we work
AI in Practice Part 2: Exploring Five Key Themes for AI Adoption at Jisc – Artificial intelligence– A follow-up blog on our AI impact workshops with
Find out more by visiting our Artificial Intelligence page to explore publications and resources, learn more about our communities and sign up for our AI Literacy training.
For regular updates from the team sign up to our mailing list.
Get in touch with the team directly at AI@jisc.ac.uk