Sydney, Jul 9 (The Conversation) "And do you work well with AI?" As tools like ChatGPT, Copilot, and other generative artificial intelligence (AI) systems become part of everyday workflows, more companies are seeking employees who can confidently say "yes" to this question. In other words, they want individuals who can effectively prompt, think alongside AI, and use it to enhance productivity.
Presently, in a growing number of roles, being "AI fluent" is fast becoming as vital as being proficient with office software once was. We've all experienced asking an AI chatbot a question only to receive a response that feels generic and surface-level. The issue isn't with the AI itself – it's that you haven't given it enough to work with.
Consider this: During training, AI systems have "read" virtually everything available on the internet. Being prediction-based, they provide the most probable, common response. Without precise guidance, it's akin to walking into a restaurant and vaguely asking for something good. Chances are, you'll end up with the chicken.
The key to maximizing AI's potential lies in understanding its adaptability to context— which you must provide. So, how do you accomplish this?
Crafting better prompts
You might have heard of "prompt engineering," which may seem like crafting a technical script is necessary for results. However, today's chatbots excel in human conversation. The format of your prompt doesn't matter as much as the content.
To optimize your AI interactions, it's crucial to communicate a few basics regarding your needs and preferences. Our recommended strategy follows the acronym CATS – context, angle, task, and style.
Context involves providing the background information and setting that AI needs. Instead of simply asking "How do I write a proposal?" go for "I'm a nonprofit director crafting a grant proposal for a foundation funding environmental education programs for urban schools." Share relevant documents, constraints, and specific details about your situation.
Angle (or attitude) taps into AI's strength in role-playing and perspective-taking. To avoid neutral responses, specify the attitude you need. For example, say "Act as a critical peer reviewer and pinpoint weaknesses in my argument" or "Adopt the perspective of a supportive mentor helping me refine this draft."
Task defines what you specifically want the AI to accomplish. Asking "Help me with my presentation" is vague, whereas "Suggest three ways to make my opening slide more engaging for an audience of small business owners" is actionable.
Style leverages AI's ability to adapt to various formats and audiences. Specify whether you need a formal report, a casual email, bullet points for executives, or a teen-friendly explanation. Instruct the AI about the voice you prefer—say, formal academic, technical, engaging, or conversational.
Context is crucial
Besides crafting clear prompts, focus on managing the surrounding information—"context engineering." Context engineering involves considering the environment and information AI accesses, like its memory function, instructions leading up to the task, prior conversation history, documents you share, or examples of ideal output.
Approach prompting as a conversation. If the first response doesn't satisfy you, push for more, request changes, or offer clarifications.
Don't expect ready-made responses. Instead, let AI trigger your own ideas. If you find valuable material but feel stuck, copy the best parts into a new session and ask it to summarize and continue from there.
Retaining your discernment
A word of caution: don't be swayed by these chatbots' human-like conversational abilities.
Always maintain a professional distance and remember you are the only thinking entity in this dynamic. Ensure to verify the accuracy of AI-generated content, as errors are increasingly common.
AI systems are incredibly capable, but they need your human intelligence to bridge the gap between their vast generic knowledge and your unique situation. Provide them with sufficient context, and they may surprise you with their usefulness. (The Conversation) NSA NSA
(Only the headline of this report may have been reworked by Editorji; the rest of the content is auto-generated from a syndicated feed.)
