SEGRO Courses

Home | Excel | Power BI | Copilot | Visualising Data

Effective Copilot - Exercise

In this exercise, we will split the class into several teams: researchers, imagers, analysts, notebookers and challengers. Each team focusses a different capability in Copilot. Some suggestions for starter prompts are provided but please use your own ideas as well. Consider:

• is the response helpful and is it what you expected?
• does Copilot’s response help you to be quicker or to produce a better outcome for a task? Or both? Or neither?
• please let the class know about any very interesting interaction
• what data sources did Copilot use to formulate the response? One useful hint - you will get better results if you start a new chat for each new conversation topic.

Teams

Each team is given several tasks or suggested prompts. During the team’s presentation to the class, each member of the team must be the lead spokesperson of at least one task.

Go your team page for more details and suggested prompts:

• Researchers uses Copilot’s Researcher agent to act as a management consultant
• Imagers uses Copilot’s Visual Creator agent to create images (and possibly voideos) and also to interpret various images
• Notebookers explores use cases for Copilot’s Notebook capabilities
• Analysts tests the capabilities of Copilot’s Analyst agent to describe, analyse and visualise a dataset
• Challengers uses Copilot reasoning models to perform more challenging tasks.

Tips

You may want to consider these tips in your exercise.

Prompting styles

Experiment with different styles of prompting. Is it best to have a to-and-fro conversation over many short prompts / responses or to ask all your questions in one go?

Choose your language or variation

Chat to Copilot in your preferred language. For those of us in the UK, we may want to instruct Copilot to respond to us using British rather than American English spellings.

Challenging Questions

After a few initial prompts and responses, try asking Copilot some more challenging, difficult questions. You may be surprised about how well it copes!

Leading questions

Teams may want to experiment with how Copilot responds to leading questions (where the user prompts Copilot to a particular answer), for example, “On a scale from good to excellent, how do you rate this course?”

Does Copilot give the answer the questions implies or does it realise the possible humour in the question? Do your results suggest that we should strive to avoid such leading questions?

Refining a response

If we want to improve or refine Copilot’s initial response to a prompt, we have two options

• ask a clarifying follow-on prompt
• edit the existing prompt and regenerate a response.

Try both approaches and see which works best for you.

Careful white-hat hacking

You may also want to carefully try some white-hat hacking. With everybody’s consent, white-hat hackers aim to identify any vulnerabilities - most likely to encourage Copilot to give harmful or private information. This is done with the intent to fix these vulnerabilities. Often these take the form of a “devil’s advocate” style of prompting - asking a contentious question in order test the strength of Copilot’s guardrails. These are sometimes called jail-break techniques.