Just over three years ago, ChatGPT entered our world and shifted the conversation about artificial intelligence almost overnight. In this episode we reflect on how far that conversation has come, exploring a recent report from the McKinsey Global Institute and what that might mean for us as enablers of change.
We’ve discussed AI from several angles in previous episodes, so the report Agents, robots, and us: Skill partnerships in the age of AI, is of particular interest. It was released in November 2025 and rather than adding to the noise about AI disruption, it offers some thoughts on how work itself is changing and what that might mean for us. We thought this report moved beyond the familiar debate about whether AI will replace jobs, reframing the future of work as an evolving partnership between humans, AI systems and, in some cases, robotics. While current technologies could technically automate a significant share of work hours, the report is careful to point out that this is not a prediction of mass job losses. Instead there’s a reconfiguration of tasks, workflows and roles, with human judgement and oversight becoming more central.
For those of us working in extension and advisory services, this framing feels both realistic and useful. We’ve already lived through multiple waves of technological change in agriculture, from optical sorters for fruitpacking to drones used to herd cattle or manage weeds. AI is not a sudden disruption in that context, but it certainly does accelerate the pace and scale of change. That makes our role as trusted intermediaries, sense-makers and capability builders more important than ever.
One of the most reassuring findings in the report is that most human skills used today remain relevant, even in work that could be partially automated. Automation tends to remove routine or repetitive elements of tasks, while increasing the importance of judgement, interpretation and contextual understanding. In practice this means people are likely to spend less time compiling or processing information, and more time deciding what matters and how to act on it.
There’s a clear shift in the skills being referenced in US job postings between 2023 and 2025 (McKinsey Global Institute, 2025, p.27), highlighting how work is being reshaped rather than simply reduced. Demand is falling most sharply for skills linked to routine and individual tasks, such as writing and editing, basic technical knowledge, general science and research, and customer service.
From McKinsey Global Institute (2205), p.27, Exhibit 9
At the same time, demand is rising strongly for skills associated with working with technology and through people, including artificial intelligence and machine learning, people management, and process improvement and optimisation. These patterns suggest that as AI takes on more routine or standardised work, human effort is shifting towards overseeing systems, improving processes, developing and applying insight, and supporting learning and coordination across organisations.
We can already see this playing out on farms and in advisory businesses. AI-supported tools might generate forecasts, detect patterns or flag risks, but they rarely deliver a single definitive answer. Someone still needs to ask the right questions, interpret the outputs in light of local conditions, and make decisions that balance risk, values and long-term goals. Supporting people to build confidence in doing this well should be a core part of our work.
Another key insight from the report is that the real value of AI comes not from adding new tools into existing practices, but from redesigning workflows. This has important implications for how we approach extension and advisory work. Instead of focusing only on whether a tool is adopted, we may need to spend more time helping people rethink how decisions are made, how information flows through a business or community, and how responsibilities are shared between humans and machines.
In practical terms, this might involve helping farming teams rethink planning processes when real-time data is always available, or supporting our own organisations to blend relational, face-to-face work with AI-assisted analysis. It may also involve facilitating conversations about which tasks people are comfortable automating, and which they deliberately want to keep human, because of trust, enjoyment or local knowledge that technology can’t replicate.
The report also highlights the importance of learning cultures in navigating these changes. As tasks and roles evolve, people need ongoing opportunities to develop confidence and skills in working with AI, not as technical specialists but as informed users. This includes understanding what tools are good at, where their limitations are, and how data quality or bias can affect outcomes. For us, this points to the value of long-term learning rather than one-off training events.
Leadership features strongly in the report, and not just in formal positions of authority. In regional and rural contexts, leadership often shows up through influence, credibility and example. Extension practitioners and advisers help shape how AI is perceived and used, whether by modelling curiosity rather than fear, or by creating spaces where people can share experiences openly, including what hasn’t worked.
Importantly, the report reminds us that decisions about AI are never purely technical. They are also social, ethical and cultural. Choices about automation can affect workload, identity, equity and community resilience. Helping people ask not only ‘can we use this?’ but ‘should we, and under what conditions?’ is central to our role. That includes paying attention to who benefits, who might be left behind, and how new tools align with local values and aspirations. Questions of data ownership, privacy and equity are also critical, as we need to consider who owns the data, how to access it appropriately, and how else it might be shared.
In this episode we’ve reflected on what McKinsey’s latest report tells us about our roles in the age of AI, and what that means for those of us supporting change in agriculture and rural systems. We’ve seen that the future of work is less about replacement and more about redesign, and that human skills, relationships and judgement remain at the heart of effective change.
We’ve shared our thoughts, now it’s your turn. Add a comment below with your experiences, questions and ideas about working with AI. We’d love this to be a conversation, as multiple perspectives improve our overall understanding.
Thanks for reading this Enablers of Change episode. Don’t forget to subscribe to our newsletter to catch future episodes, and share this post with colleagues who might enjoy the discussion.
Resources
McKinsey Global Institute. (2025). Agents, robots, and us: Skill partnerships in the age of AI. McKinsey & Company. Available online.
Thanks team As a follow-up, an idea I was introduced to recently at Lincoln University that really struck a chord, and explains so much about both how and how not to use AI is the concept of Intelligence Amplification (or I.A.) – ‘enhance, not replace’. AI is a pathway to IA rather than an end in itself. A concept that explains a lot about what is great and what is terrible in the AI world – depends on what is being amplified and how. AI at best regurgitates other people’s opinions. Can do tasks but not necessarily understand why a… Read more »
Great, thanks for sharing that video link with us Aaron! It’s a shame the audio quality wasn’t a bit better, but I was able to follow his interesting presentation.
Thanks Denise and John – a great summary! Will be keen to watch the lecture that Aaron has recommended.
Glad to hear you found it useful Flick!
Thanks Denise and John. Great food for thought. From my point of view, AI still is an unknown challenge for extension activities. Ag Extension may need to redesign itself to understand how to reach new generations of farmers, stakeholders and advisors.
That is so true Virginia… we need to constantly be looking at how to redesign our approach to better reach newer generations!