The world have suddenly woken up to the possibilities and dangers of AI, and the sphere of education is no exception.
There have been breathless articles about the ‘End of Education’ as the fear mounts that learners can now hand in ready-made essays. This is not the end of education. It is the end of education as we know it, and that should have happened a while back.
Still the dawn of the web, some learners have been handing in work that is remarkably similar to the Wikipedia page on that topic. The current level of consumer AI moves us a little further down the same path. But the revolution has not arrived… yet.
Right now, the results of OpenAi via Microsoft’s Bing, or Google Bard are impressive compared to what’s come before, but still not great. When you grew up with the Microsoft Office paperclip as your go-to reference of computer intelligence, you will no doubt find the progress staggering. But with the latest public AI systems, there are still often mistakes and errors which a little fact-checking can uncover. But it is only a matter of time before AI offers a completely convincing and accurate ability to write essays.
There will be a range of ideas from teachers about the point of education, but if part of the purpose is putting needed facts at the fingertips of people when they need them, and the skills in their fingers (and elsewhere) to utilise them in creative ways, drawing on wide, diverse experiences, AI provides opportunities to educators. But it does mean that the way we teach and assess has to change.
If knowledge acquisition is your goal for your learners, assessment needs to move to a more personable model. Just like a PhD candidate defending their work, the assessor will need to sit down with the learner an (nicely) interrogate what they know. Discussion will need to replace the discussive essay. Presentations, group work and practical doing will need to become the new gold standard of formative assessment, as it largely has already become beyond the education silo.
At the moment there seems to be an arms race to try to beat AI. For example, the popular University Plagiarism software TurnItIn has announced that it has released a tool to spot AI generated content, and many schools are likely thinking about banning AI generated work too. But should AI be banned, or should institutions, including schools look to adapt and harness the Jinn now it is out of the bottle? The use of AI will be pervasive in the future, and it is the educator’s duty to help learners make the most of any new technology, as well as building the skills and strategies to mitigate harms.
Like the trickster Jinn who twists every wish into something unintentionally awful, the prompts we give to the current crop of AI platforms available can lead to a masterpiece, or a piece of something completely different. Just like writing code, which is essentially what you are doing with inputs via a natural language engine, the way an input is given can result in errors and nonsense. Honing the prompts given to AI platforms is something that teachers can help learners do better to get a desired result.
I am excited by the ways that even at this early stage, AI platforms can be used in lessons today to help our young people learn. From offering a foreign language partner to ‘text’ with live feedback, or creative storytelling where learners construct a story together. Perhaps the most exciting is the new opportunities AI can offer to those differently able to carry out tasks in the classroom many of us take for granted.
Generative videos are already here, where videos can be created from nothing more than a few typed words. True, most currently look like a mad kind of fever dream, but to far in the future it is easy to see videos or VR experiences being automatically created from nothing based of the feedback of what your learner got wrong or right in class. Metaverse schools, anyone?
But it is important not to have a blinkered view. In the near term, there is concerns about the data we all provide to these platforms and what may be done with it. Things are likely going to get much worse in the way of fake news and information overload while we become used to dealing with AI in our daily lives. In the long term, it is very likely that AI will make many of the things we do as teachers obsolete. You can either try to hold back the inevitable tide in which your learners will surely have to swim in, or you can do what good teachers do best, learn, adapt and use new advances available to you to remain relevant and provide the best education for your learners with a little help from our robot assistants.
ICTmagic 