Are you skilled in utilizing AI?
As tools like ChatGPT, Copilot, and a variety of generative artificial intelligence (AI) systems become part of everyday workflows, companies are increasingly looking for employees who can confidently reply “yes” to this query. In other words, they need individuals who can adeptly prompt, collaborate with AI, and use it to boost productivity.
ADVERTISEMENT
CONTINUE READING BELOW
Read: AI boosts Sars efficiencies, states Kieswetter at G20 Zimbali meeting
Indeed, in a widening range of positions, being “AI fluent” is swiftly turning into a necessity, much like proficiency in traditional office software once was.
Nevertheless, we’ve all encountered that annoying moment when we ask an AI chatbot a question, only to receive a response that feels too generic or shallow. The problem isn’t with the AI; it’s that you may not have provided enough information.
Consider this: during its training, the AI “read” nearly everything available online. However, since it operates based on predictions, it frequently delivers the most typical or common responses. Without specific guidance, it’s similar to entering a restaurant and asking for something delicious—you’re likely just going to end up with chicken.
The key is to realize that AI systems excel on context, but it’s up to you to supply that. So, how do you achieve this?
Creating Enhanced Prompts
You might have heard of the term “prompt engineering.” It may seem like you need to create a technical script to obtain the desired results.
Fortunately, today’s chatbots are quite adept at human interaction. The structure of your prompt isn’t as crucial as its content.
To make the most of your AI interactions, it’s essential to communicate several key aspects about what you want and how you want it. We suggest using the acronym CATS—context, angle, task, and style.
- Context involves providing the background and setting the AI needs. Instead of asking, “How do I write a proposal?” you could say, “I’m a nonprofit director drafting a grant proposal to a foundation that supports environmental education programs for urban schools.” Upload relevant documents, share your constraints, and outline your specific context.
- Angle (or attitude) taps into AI’s ability to role-play and embrace various perspectives. Rather than seeking a neutral response, specify the tone you want. For example, “Act as a critical peer reviewer and identify weaknesses in my argument” or “Take the perspective of a supportive mentor helping me improve this draft.”
- Task refers specifically to what you want the AI to do. “Assist me with my presentation” is vague. In contrast, “Provide three suggestions to make my opening slide more engaging for an audience of small business owners” is much more actionable.
- Style makes use of AI’s ability to adapt to different formats and audiences. Specify if you need a formal report, a casual email, bullet points for executives, or explanations suitable for teenagers. Clarify the tone you want to employ—formal academic, technical, engaging, or conversational.
Context is crucial
ADVERTISEMENT:
CONTINUE READING BELOW
In addition to crafting a clear and effective prompt, consider managing the overall information—this is what we define as “context engineering.” Context engineering includes everything surrounding the prompt.
This involves pondering the environment and information available to the AI: its memory function, preceding instructions, conversation history, documents you upload, or examples of what quality output looks like.
Think of prompting as a dialogue. If the initial response doesn’t fulfill your expectations, ask for more, request modifications, or provide additional context.
Do not expect the AI to supply a fully formed answer. Instead, view it as a catalyst for your own thought process. If you find that the AI has produced valuable content but you’re unsure about the next steps, extract the best parts and transfer them to a new session, asking the AI to summarize and continue from that point.
Staying Sharp
A note of caution: Don’t get too caught up in the conversational capabilities of these chatbots.
Always keep a professional distance and remember that you are the only thinking entity in this interaction. Furthermore, ensure to verify the accuracy of any information the AI generates—errors are becoming increasingly common.
AI systems are incredibly capable, but they depend on you—and your human intellect—to bridge the gap between their broad general knowledge and your specific situation. Provide them with enough context, and they might surprise you with their helpfulness.
Sandra Peter is a Director of Sydney Executive Plus Business School, University of Sydney; and Kai Riemer, Professor of Information Technology and Organisation, University of Sydney.
This article is republished from The Conversation under a Creative Commons license. Read the original article.
