Anil Khamis & Zeenar Salim
With the introduction of the OpenAI ChatGPT in November 2022, there has been a fury of activity accompanied by doomsday predictions for the future of humanity.
What is ChatGPT, what are its capabilities, and why is this remarkable outpouring of predictions of job losses, replacement of teachers, and hyper-accelerated automation?
ChatGPT is an online, open-source application that uses the piles and piles of data in the digital ecosystem to ‘write’ text, codes, formulae, and the like that ‘sound’ human-like. It is one of the most downloaded and used applications with over a 100 million users who are students writing essays for their courses, professionals using it for computer programming coding, researchers to generative survey questions, and the list goes on and on.
ChatGPT and other applications, such as Google’s Bard, are programmed with algorithms that are a set of rules to process information or data. These algorithms ‘recognise’ patterns – the value of AI tools is that they can go through large datasets in a very short time. So it appears that the AI can almost instantaneously ‘write’ what you ask it to based on the ‘prompt’ provided.
Two things to be clear about: (i) the AI is not ‘thinking’; it is just re-producing text based on pre-existing information that is provided by humans, therefore, one cannot trust the information that has been generated, unless an established authority authenticates it and (ii) it does not have moral and ethical reasoning implying it does not ‘know’ what is right or wrong. It just mimics what has been programmed for it.
So, how do we use AI then?
What are its implications for education, then? ChatGPT3 can be used as a tool to spur thinking, however, it should not be used as an authority on the generated information. In addition, one cannot trust AI for the authenticity of the information it produces, implying that it acts as a stochastic parrot that uses patterns of previously available data to ‘generate’ original-looking text, code, or images.
While students across the planet, are using this application, it is important to raise awareness of the silver lining in the use of AI. There are two important hazards in unthinking use of AI generated information. These can have profound effects on the quality of education at the school and higher education level.
The last human chess champion Gary Kasparov is instructive in this area. He helped IBM develop ‘Deep Blue’, a chess programme that ultimately defeated him. He asserts that we must work with AI and machines learning in a collaborative manner – it notes that human creativity can flourish with AI and machine learning capabilities. Today, chess programmes compete against each-other and human cannot compete with such computer processing power that can review trillions of options in split second time frames. However, they don’t enjoy competition, have fun, or get a thrill from the win or the disappointment of a loss.
Humans on much less information with their powers of imagination, inference, moral reasoning, and experiential learning are far more creative, intuitive, and we might say, wise.
How do we then prepare students to use AI in education? Descriptive writing is well written by the AI.
Compelling narratives sharing cultural stories are not. How do we engage students in telling authentic and creative stories?
How do we engage students in inquiring and co-creating foundational knowledge instead of reproducing of factual knowledge? Examples include asking students to engage in a community service project and writing opinion articles for a local newspaper on their experience and learning from community service projects.
In the case of mathematics, how do we move away from memorization of formulae and procedures, to engage students in an explanation of the concept at hand and make estimations.
Holistically, focusing on the core goal of education, we can decide if or not AI can be used for a particular project and help students also raise that awareness. The questions that we need to ask ourselves, are how do we use AI to enhance students’ skills such as logical reasoning, decision-making, intuition, moral reasoning, and leadership?
As educational systems confront the challenges of AI, we cannot afford to use such tools without thinking, guidelines, and understanding their uses. Such a scenario will lead to the further ratification of education without any beneficial outcomes. With considered use and application, we need to perhaps, move the needle from teaching descriptive writing to teaching narrative writing rich in meaning situated in the context, engage whole new processes of learning, discovery, and time-saving opportunities arise to free learners and their educators to explore new vistas.
We need to research these new technologies and ‘control’ them for our benefit.