Is the ChatGPT the new calculator

The image above is produced with another OpenAI tool DALL-E https://openai.com/product/dall-e-2 

OpenAI's chatGPT is a conversational piece of software that can also produce answers (or tries to) based on what you ask. So if you asked it to "Write a descriptive text on the role of chaptGPT in the workplaces", it will. An example of this, using, at the time of writing the latest version using GPT4 via https://poe.com/gpt-4 is shown below 

It is good and so has rapidly produced debate on its impact. But is this debate new?

When calculators became affordable enough that it become reasonable to factor into assessed work at school and universities the was debate that it might be cheating to use a calculator. We are perhaps at a similar stage now with chatGPT is it cheating to use it. There are perhaps lessons to be learnt from introducing calculators to assessments and in the workplace.

It is cheating?

Is it, or is it cheating for what is currently being done? As an example if you are testing someones ability to do mental arithmetic then a calculator would appropriately be seen as cheating and so you wouldn't allow a calculator. The first question though has to be what are you assessing or in the workplace needing the person to do though, if they are applying techniques to solve a problem then maybe it is not cheating - the calculator is an aid. Is that how we should look at chatGPT as an aid to us, possibly enabling us to focus on the task? Maybe. This view has been by a number of people.

In the Bloomsberg article (Aldrick, 2023) this argument was put forward and it aligns with the view of others (e.g. Feretzakis et al, 2020) that Artificial Intelligence techniques can be used to complement 'human' work. The jury is out and will be for a very long-time on this.

Are there opportunities here?

So let us take the lesson learned from calculators in education, what happened, the assessments changed with arguably deeper consideration of what was need to be assessed and why. Could the same happen with chatGPT I think yes, these chatbots can produce some very good starting points for an assessment on an essay and chatGPT can produce simple programs. So perhaps assessments need to change be more personalised or make use of it with assignments? Though not specifically focussed on chatGPT but looks at authentic assessments might this help?

What about work. This is always a controversial one but some jobs may change but new ones arise and other opportunities. Might it change further how articles are produced?

In the image below some of the text from this blog was put through the more widely used chatGPT (using GPT3). I think it lost a few of the points I wanted.

References

Aldrick, P. (2023) ‘ChatGPT Will Be the Calculator for Writing , Top Economist Says’, Bloomberg, pp. 1–2. Available at: https://www.bloomberg.com/news/articles/2023-01-18/chatgpt-will-be-the-calculator-for-writing-top-economist-says.

Feretzakis, Georgios, Evangelos Loupelis, Aikaterini Sakagianni, Dimitris Kalles, Maria Martsoukou, Malvina Lada, Nikoletta Skarmoutsou, Constantinos Christopoulos, Konstantinos Valakis, Aikaterini Velentza, Stavroula Petropoulou, Sophia Michelidou, and Konstantinos Alexiou. 2020. "Using Machine Learning Techniques to Aid Empirical Antibiotic Therapy Decisions in the Intensive Care Unit of a General Hospital in Greece" Antibiotics 9, no. 2: 50. https://doi.org/10.3390/antibiotics9020050

All opinions in this blog are the Author's and should not in any way be seen as reflecting the views of any organisation the Author has any association with. Twitter @scottturneruon