{"id":200489,"date":"2025-10-09T16:09:08","date_gmt":"2025-10-09T16:09:08","guid":{"rendered":"https:\/\/www.newsbeep.com\/ca\/200489\/"},"modified":"2025-10-09T16:09:08","modified_gmt":"2025-10-09T16:09:08","slug":"ai-requests-should-begin-with-introducing-yourself-and-then-offering-a-long-query","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/ca\/200489\/","title":{"rendered":"AI requests should begin with introducing yourself and then offering a long query"},"content":{"rendered":"

Interested in more careers-related content? Check out our new weekly Work Life newsletter<\/a>. Sent every Monday afternoon.<\/p>\n

Prompting is a new skill that most of us will need to learn to be proficient at our jobs. It\u2019s not like searching on Google, even if artificial intelligence\u2019s large language models are a technology with an inviting search box. <\/p>\n

Instead of just typing in a quick query, tech consultant Rajeev Kapur suggests beginning any complicated search by introducing yourself to ChatGPT or its equivalent technology. It can\u2019t read your mind, lacks life experience, and will take your words very literally. So tell it about yourself in a couple of sentences and then it can gauge better what you might need. <\/p>\n

\u201cWhen people get bad results from a ChatGPT request, it\u2019s often because they are feeding it short, poorly written prompts and expecting the technology to magically \u2018get\u2019 what they\u2019re after,\u201d he writes in Prompting Made Simple.<\/p>\n

\u201cOn average, a single good-to-great prompt can be anywhere from 100 to 400 words, so don\u2019t limit yourself.\u201d <\/p>\n

ChatGPT has a memory function and will start to know you if you turn that feature on in the personalization settings. But you still want to prompt fully and also assign the chatbot a persona so it will be better able to tailor its responses. For example, if you ask for sales tips you\u2019ll get basic, vanilla stuff. Instead, he advises asking it to assume the role of an expert salesperson talking to another expert salesperson, which will unleash more advanced advice. <\/p>\n

\u201cYou can literally ask ChatGPT to be anyone in the past or present, real or fictional, and then have a two-way conversation with the persona. Try it!\u201d he says.<\/p>\n

He adds that it\u2019s important to provide context; not be afraid to reword your query and ask again in a different way; and to instruct the technology to ensure your answer is unbiased and does not rely on stereotypes if you think that may be an issue. <\/p>\n

Set constraints and format: You may want your answer in the form of a LinkedIn post, a press release, a formal letter or some other type of writing; you also may want to specify tone and length. Contrary to what you may have heard, he insists there is no need to be polite, saying \u201cplease\u201d or \u201cthank you\u201d or even \u201cI would like you to.\u201d It doesn\u2019t care. <\/p>\n

\u201cClarity and specificity are key. A well-crafted prompt guides the AI to provide the most accurate, relevant and helpful response,\u201d he notes. His guiding formula is RTCA, for role, task, context and ask: Assign it a role; introduce a task to perform; explain any context that is important; and ask it to obey any criteria you feel necessary. <\/p>\n

You also may want to chain your prompts, breaking them down into smaller, interconnected ones:<\/p>\n

In \u201cchain of thought\u201d prompting, you ask the chatbot how it came to that answer because that might help you solve similar problems. He offers this example: What were the causes of the First World War; explain step by step.\u00a0 In \u201cfew-shot\u201d prompting, you provide it with some examples of a task along with your request. You might, for example, give it a capsule marketing description of a smartphone and a wireless earbud and ask for a similar version for a new smartwatch.In \u201cgoal-driven\u201d prompting, you specify objectives and parameters. For example: My goal is to teach Grade 8 students about the water cycle in an interactive, engaging way. Create a lesson plan with an introduction, hands-on activity and a closing discussion.In \u201cexploratory\u201d prompting, you encourage it to discover various perspectives and possibilities. For example: If our business wanted to expand into a completely new industry, what innovative products or services could we offer that fit our strengths? In \u201cverification\u201d prompting, you ask it to check its work. For example: Review your explanation and check whether the assumptions are supported by evidence. If not, rewrite your report with stronger justification. <\/p>\n

As we teach ourselves to use large language models, those are some lessons to keep in mind. <\/p>\n

Quick Hits<\/p>\n

Ottawa productivity coach Chris Bailey says in his newsletter<\/a> when he is tempted to procrastinate he gives himself a choice: Either do the task now or write down in his journal why he wants to put it off. Most times, that choice overcomes his aversion to getting started. Author Mark Manson notes<\/a> the best investments are the things no one can take away from you such as knowledge, experience and \u00a0mindset. The weakest investments are things you can lose at any time such as status, emotions and material things. \u201cConsistency enlarges ability,\u201d says<\/a> Atomic Habits author James Clear. <\/p>\n

Harvey Schachter is a Kingston-based writer specializing in management issues. He, along with Sheelagh Whittaker, former CEO of both EDS Canada and Cancom, are the authors of When Harvey Didn\u2019t Meet Sheelagh: Emails on Leadership.<\/p>\n","protected":false},"excerpt":{"rendered":"Interested in more careers-related content? Check out our new weekly Work Life newsletter. Sent every Monday afternoon. Prompting…\n","protected":false},"author":2,"featured_media":200490,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[62,276,277,49,48,4081,61],"class_list":{"0":"post-200489","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-ca","12":"tag-canada","13":"tag-ordid20000","14":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/200489","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/comments?post=200489"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/posts\/200489\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media\/200490"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/media?parent=200489"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/categories?post=200489"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ca\/wp-json\/wp\/v2\/tags?post=200489"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}