This is my penultimate blog for the Self-Improving System project. A great deal has changed since I embarked on this project in May 2022. We have had very considerable political turmoil, of course, that has hit the education department harder than most. Probably more significantly, we have had the ONS reclassification of colleges as public sector.
Without implying any criticism or blame towards any official – very sincerely – I consider the reclassification process to be a symptom of collective incompetence in Whitehall. It is a classic example of why it is wrong to imagine that all that needs to happen to improve the lot of FE is for some key powerful individuals to ‘get it’ – a minister, perhaps, or the officials at Treasury in the education spending team. Nobody particularly wanted the ONS decision, and nobody intended for it to be a key driver of change in the sector. And yet this is now happening, because being a system does not imply being a controlled system.
Many of my interlocutors have misunderstood this point, saying things like “FE isn’t a system because it isn’t managed as a system”. This is a failure of insight. The planet earth is a system but isn’t managed. My house has a heating system, but I’m not managing it; I have certain inputs into it, but I don’t understand how the whole system works and I’m not managing the whole thing, I’m simply influencing it, sometimes intentionally and sometimes in ways I don’t intend and with consequences I don’t want. And so it is with FE – it is a system that we must map and understand as a system if we are then to collectively make the changes to system design that will lead to better, more sustainable and self-improving behaviour and outcomes.
But in the wider world there have been bigger changes. Obviously international events continue to have powerful and complex knock-on effects for FE. But we have also seen the arrival of the first of a new generation of completely open and accessible AI tools in ChatGPT. We are all still grappling with how this changes the worlds of work and education, but it has a direct and surprising effect on my own project.
Very early in my discussions with the ETF and colleagues at Saïd I realised that there would be two types of outcome for my project, and both would be important. The first would be some form of written artifact: the summation of 10 months of work on the question “what might be the features of a self-improving system in FE, and how can we get there?” But there would be a second kind of outcome, which would be the new thinking of the participants generated by the work, the new patterns of collaboration, new relationships, new tools being deployed by key people across the system to forge a collective understanding of what change could mean, and to start the long process of working together to effect those changes while still doing their day jobs within the system.
ChatGPT has made me realise that of these two types of outcome, the second is by far the more important. Why? Simply because, in 30 seconds it can give a pretty good answer to the questions at the heart of my work. Not a definitive answer. Not the best answer we can produce. But an answer that is certainly good enough to act as a simple instruction guide to make great progress.
But producing a good answer doesn’t change anything in the world. It is only if the people in a position to lead change are bought into that answer and resolve to move their behaviour in the directions it suggests that change will happen. And that in turn will only happen if people have been involved in the work and have confidence that it is the collective thinking of a wide enough set of high-calibre contributors who are well-placed to do the work.
That is why this year has made clearer than ever that the impact of a project like this lies only marginally in the actual answers produced, and instead lies greatly in the far less tangible outcomes of the momentum and engagement created by it, the collegiality, the sense of collective endeavour to come together and keep doing the hard work that will improve the world if we grapple with it in an open, sustained, collaborative and determined way.
My final blog will signpost both outcomes from the project – the “answer” to the exam question, and the more important impact of how the work has been done.
Meanwhile, over to ChatGPT:
“A self-improving FE (Further Education) system in England would likely have several key features, including:
“Moving from the current state of the Further Education (FE) system in England to a self-improving system will require a coordinated effort from multiple stakeholders. Here are some steps that could be taken to facilitate this transition:
Overall, moving towards a self-improving FE system will require a long-term commitment from all stakeholders, as well as significant investment in infrastructure, technology, and professional development. However, the potential benefits for learners, educators, and the wider economy make this transition an essential one for the future of Further Education in England.”
“The role of the Government in supporting the reform programme for a self-improving Further Education (FE) system in England is critical. Here are some ways the Government can support this reform programme:
Overall, the Government plays a vital role in supporting the reform programme for a self-improving FE system in England. By providing funding, setting policy direction, ensuring accountability, facilitating collaboration, supporting research and development, and communicating the value of Further Education, the Government can help to create a system that delivers high-quality education and supports the needs of learners and industry partners.”
David Russell
Executive in Residence at Oxford Saïd Business School
Education and Training Foundation