Reflections:

1 Sign into a text generation tool, try out the Medieval Plague simulation that Benjamin Breen put together for his students (from this module’s Read/Watch section). What do you notice about the outputs? Does it seem accurate in its depiction of the times? Based on your own gaming experience, how successful is the prompt at creating successful game mechanics?

The stimulation ‘s trials using large-scale language models (LLMs), such ChatGPT, as teaching aids in their history class at the University of California, Santa Cruz, starting in January 2023, are the main subject of this book. The stimulation started in 2019 with the usage of GPT-2 to examine the effects of AI in education, and in 2022 it concentrated on his wife’s consultancy work for OpenAI, which included GPT-4’s initial release through adversarial testing.

The author expresses the opinion that LLMs may be utilized to replicate interactive historical settings for academic projects. Although the author notes that these simulations may contain mistakes, inconsistencies, and hallucinations, she stresses that hallucinations are not always a bad thing. It may be used to replicate various historical situations; there were some early testing findings released too.

From long-term prospects, the stimulation is bullish about LLMs’ potential to improve higher education, particularly in relation to the value of the humanities. He made the point that LLMs have a tight relationship to texts and are intimately tied to the approaches and analytical techniques that are stressed in university humanities courses. According to the author, history students have already had some training on the potential issues with LLMs because they have studied how to tackle issues comparable to these in services by using techniques like audience research, genre analysis, and fact checking.

However, author also addresses potential short-term issues, particularly for teachers who may need to reconsider how assignments and homework are designed. He underlined that employing role-playing techniques and modifying instructions will make it more likely for pupils to cheat on their schoolwork. In the end, the paper presents positive opinions on the possible contribution of LLMs to education, but it also highlights several issues that require attention.

2 Where do you think these tools will be in their evolution in 2-3 years’ time?

This is a really interesting topic that has been much discussed in the media when those tools first time came out, it also started me thinking when working on this module. In my opinion, AI tools would affect people from various aspects, for example, in 2-3 years’ time, these tools could create intelligent furniture and it would become popular among a lot of families, at that time, when people go home, the lights in the room would come on automatically, the curtains will open by itself during daytime, it can even help you finish some easy works such as ordering take-out. From the education aspect which is our course mainly focus on, I think AI tools could possibly be the instructor for some class by teaching students just like the real instructors, unfortunately, Ai instructor might do better due to Ai could provide students personalized learning content and unique teaching methods according to the characteristics and needs of students, it will improve the quality and effectiveness of education and help students better develop their potential, and this really might happen in the future. 

Reference:

OpenAI. (2023, November, 26). [ChatGPT response ]. https://chat.openai.com/