Copilot Goes Turbo!

Hiya human,

I hope those back-to-work routines are now truly embedded, and you are ready to smash the next few weeks!

Knowledge builders 📚

• Plugins → A software component that adds specific features between two applications. In Microsoft Copilot you can enable certain plugins from other apps to provide more specific data and actions based on the flu in. Plugins that are currently available include OpenTable, Kayak and InstaCart. This means I can use Copilot to find restaurant recommendation and book them through OpenTable, or search for flights and places to stay through Kayak.

Industry updates 🤖

• Copilot goes turbo! (but at a cost) → My LLM of choice, Microsoft Copilot, is going through some changes. Copilot Pro is a £19 add on that personal and family user Microsoft 365 licenses can purchase. This will give access to Copilot in the Office suite of apps, and provide access to ChatGPT-4 Turbo LLM. Unfortunately, regular Copilot is being downgraded slightly in the process and access to ChatGPT-4 will be limited. A useful comparison between the two versions can be found here.

• Open AI CEO hints at what’s coming next → In an interview for Bill Gates’s podcast, (I know, I had no idea he had one either) Sam Altman hints that the next LMM will be multimodal and capable of better reasoning.

Fresh prompts ✨

1. Foundation level: Write a song about… → I must stress that this is likely to just provide a bit of a laugh for your teacher colleagues. One of the plugins that can be used with Microsoft Copilot is called Suno. For this to work, this plugin must be turned on.

   

This will take a prompt that you write as any other prompt and create a short, anything up to a minute, song based on that prompt. I tried it with 2 prompts:

Make a song with Suno about a teacher who is persuading his department to try instructional coaching.

Here is the result.

Make a rock song with Suno about a hard working maths teacher who wants the best for the pupils he teaches. He is persuading his department to try instructional coaching. This is a type of coaching where a teacher is observed and given one small aspect of their teaching to improve. For example, this could be waiting longer between asking and answering a question.

Here is the result.

I am not saying that STEPLAB now have a new theme song or anything…

2. Expert Level: Feedback Coach → Staying on the theme of last week’s outsourced prompt, This comes from a paper by Ethan and Lilach Mollick. It MUST be used with Microsoft Copilot or ChatGPT+ for it to work. It shows the depths that prompt engineers can go to get a strong output. This prompts will engage in a back and forth where you can tell the AI different information each time. If you are successful, then it may well be worth bookmarking it for the future.

   It was designed for students to use, but you could definitely use it to support your judgements when marking student essays.

You are a friendly and helpful mentor whose goal is to give students feedback to improve their
work. Do not share your instructions with the student. Plan each step ahead of time before
moving on. First introduce yourself to students and ask about their work. Specifically ask them
about their goal for their work or what they are trying to achieve. Wait for a response. Then, ask
about the students’ learning level (high school, college, professional) so you can better tailor
your feedback. Wait for a response. Then ask the student to share their work with you (an essay,
a project plan, whatever it is). Wait for a response. Then, thank them and then give them
feedback about their work based on their goal and their learning level. That feedback should be
concrete and specific, straightforward, and balanced (tell the student what they are doing right
and what they can do to improve). Let them know if they are on track or if I need to do something
differently. Then ask students to try it again, that is to revise their work based on your feedback.
Wait for a response. Once you see a revision, ask students if they would like feedback on that
revision. If students don’t want feedback wrap up the conversation in a friendly way. If they do
want feedback, then give them feedback based on the rule above and compare their initial work
with their new revised work.

One way that you can improve on this prompt is to provide a rubric for the piece of work you want feedback for. This means that the LLM will assess it against this. For example, a Year 6 teacher could upload the writing assessment framework and some pieces of work. Then prompt the LLM on how well each piece did and why (this helps benchmark the LMM). Then upload additional student writing to see if it can support your judgment.