Before using Generative AI

Generative AI tools are powerful platforms that can transform work and creativity. However, there are some really important characteristics that you need to be aware of. Tools like ChatGPT, Gemini, and Claude are specialist technologies that need to be understood before engaging too deeply with them.

Some of the inherent risks and concerns include:

• Privacy issues - make sure you have permission to upload any content to these platforms, as it will most likely end up as training data for these models.

• Inherent bias - many of these platforms produce biased results and outputs which can magnify pre-existing biases within society.

• Misinformation - these technologies do not produce factual information, but probabilistic outputs, which, for the most part, resemble factual content.

• Ethics - much of the digital information used to train these models has been stolen from authors, organisations, academics, and creative artists.

• The environment - the environmental cost of training these models and producing outputs is immense and places significant strains on already dwindling environmental resources. Use these tools sparingly.

• Cognitive offloading - an over-reliance on these platforms is detrimental to learning, reducing inherent creativity and traditional cognitive function.

Also, from an educational point of view, it is easy to be impressed with these tools and use them without considering the educational value involved (or lack thereof). With all things educational, it should be "Pedagogy first, Technology second".

• Cognitive offloading