How to update your teaching in a world with AI
AI Large Language Model tools like ChatGPT bring new opportunities and challenges to teachers and educators. To handle them, it makes sense to take a step back and explore your options.
Complex problems are often vague and have many possible solutions. The Meta-Problem Method may lead you far away from the dilemma that started your quest. That’s because the method forces you to clarify what you really want and what you are willing to give up. It enables you to compare objectively the possible pathways and their trade offs. It prevents you locking into solutions mode too early and then doubling down on solving a low-yield problem that does not serve your goals as well as the alternatives. At the end of this process, you will have a better understanding of your priorities and how to achieve them.
Steps in the Meta-Problem Method
Dilemma
The high-level issue you are trying to address
Use new technology to improve students’ learning.
Goal
The changes you want to make to address the dilemma. There are usually many options.
Supporting goals
- Avoid students choosing to cheat.
- Teach students how to use new technology.
Other goals could include students learn the course content or reducing teacher work.
Problem Space
The set of problems you could chose to solve to advance your goals, plus the constraints that hold you back.
Example problems:
- How can we prevent students cheating with AI? Maybe the problem to solve is “how can we discourage cheating with solutions like making it harder to do?”
- How can we ensure students learn how to use new technology? Maybe the problem to solve is “how to teach students when AI gets the answer wrong?”
There are many other potential problems to solve related to using new technologies to improve students learning. Each goal has many possible problems we could link to it. Are there other problems linked to these first two goals? Which options come to mind for the other goals?
High-Yield Problems
Sometimes solving one problem helps make progress towards several goals. In this step, we identify these “two-for-the-price-of-one” problems.
Which options will advance more than one goal?
- Switching to blue-book tests ensures students learn the material and prevents them from cheating. However, it would not help them learn when to use AI.
- Designing course assessments to test the things AI does both well and poorly will help them learn how to use new technology and some of the course content. However, it increases the risk of students using AI inappropriately.
There are many potential solutions that will have varying effects on the set of goals. Which alternatives improve the most important goals? How might the unknown change the right path forward? What other possible solutions are there to address the dilemma?
Problem Selection
Which of the many possible options in the high-yield problem step is the best set to address the dilemma?
- Which solutions make the most sense as an educator?
- Which solutions will best address the dilemma?
- Which solutions will deliver the best outcome for the least amount of time, effort and money?
Implement, Learn and Adapt
Check continuously that you are still solving the best problem, as new information emerges.
Observe and learn as you go. As new information reveals itself, check continuously that you’re still solving the right problem.