Dortch’s WAHAT #12: Have You Conversed With Your AI Today?
In the “M.A.S.H.” content universe (books, movie, and TV shows), my favorite character was and is Corporal Walter Eugene “Radar” O’Reilly. He’s right up there with the original S’chn T’gai Spock from the “Star Trek” content universe. In both cases, I remain torn between wanting to be or be like that character and wanting to have 24-by-7-by-365 access to one or the other on demand. (And for those of you who may be wondering or concerned about me, I got both full names online and not from memory.)
The real world can now approximate that second desire of mine for almost anyone with online access. And the role of a nearly perfect on-demand assistant with just about anything is one that artificial intelligence (AI) can play in ways that enhance almost any human activity and humans’ satisfaction with that activity, without eliminating or demeaning a single job or person.
To take full advantage, however, each of us will have to rethink how we approach something most of us probably think we know all we need to know. That something is how we search for information online.
Online Search: No Longer a Task, But a Conversation
You may have already noticed that you’re encountering terms such as “search queries” and “search keywords” less and less frequently, while encountering terms such as “AI prompts” and “AI prompting” more often. Here’s a brief, clear description of the difference between the two, generated by Google AI itself.
“Search queries are keyword-based inputs for traditional search engines to find relevant web pages, while AI prompts are natural language instructions, commands, or questions given to generative AI models, requiring more context and detail to produce targeted, synthesized, or creative outputs.”
This means we need to treat our interactions with AI-powered research tools more like interactions with humans. Those AI interactions must become more detailed, conversational, and iterative. This often begins with you asking AI an initial question, perhaps resembling a traditional search engine query. You then receive an initial answer, after which you can ask the AI for more details, or to clarify or build upon that initial response. Lather, rinse, and repeat.
A conversational approach to AI queries is far more likely to generate responses that enhance your knowledge and understanding of the initial question topic than almost any but the most complex search engine query.. It also creates more opportunities for the AI to get smarter and, where appropriate and permitted by you, more and better personalization.
Imagine the world’s most skilled librarian with knowledge of and access to all the world’s accessible publications and repositories of collected knowledge, and the world’s most amazing summarization and presentation skills. Now, imagine unfettered access to this super-librarian any time you wanted or needed to know something or more about almost anything.
The current “civilian” versions of AI services from Google, Chat GPT, and others already offer abilities and resources that point in this Radar/Spock/Super-Librarian direction. All we have to do is ask better questions and have more conversations when interacting with AI tools. And all will be well.
But Of Course, There’s a Catch…
Here’s a bit more from what Google AI has to say about the difference between traditional search queries and interactions with AI. “The key difference is that search queries retrieve existing information, whereas AI prompts direct the AI to generate new content or provide synthesized answers based on the instructions provided.”
As I imly recall, the late, great comic genius George Carlin had a sportscaster character named “Biff Burns.” His signature sign-off was, “I call’em as I see’em — and if I don’t see’em, I make’em up!”
AI is sometimes like that. Hence, the frequency with which we’re encountering terms such as “AI hallucinations.” Also, the disclaimers found at the bottom of many AI-generated responses that mention the possibility of AI-generated mistakes.
These and other shortcomings will be overcome, even as new ones appear. That’s how broadly adopted technologies and our responses to them evolve. Just like with harmful bacteria, viruses, and other threats, and the vaccines and other treatments and techniques we employ in response.
Meanwhile,…
Dortch’s Recommendations for Interacting with AI (at least for now)
- Think, then type. Frame your inquiry and resulting prompt or prompts as if you were engaging an audience or your colleagues. Most ad hoc, poorly framed AI prompts will generate results no better than those from a traditional search engine experience.
- Don’t fear questioning the answer. Be prepared to follow up on any parts of the initial answer you get that are unclear or that trigger additional curiosity. (You might want to take notes as you go, or at least capture a transcript of your conversation.
- Don’t resist having your question questioned. When AI asks you for clarification, two things happen. The AI gets a little smarter, and you get encouragement and perhaps even helpful hints to refine your prompt(s).
- Ask again. And again. Submit the same prompt to more than one AI tool. Maybe even as one or more AI tools to analyze and respond to two or more responses to your original prompt. And make it a habit to ask at least two versions of every prompt. Don’t miss any opportunities to learn or see something new, interesting, and/or useful.
If credible, useful, and verifiable knowledge matters in your life, you should begin developing conversational relationships with AI tools ASAP. Since they aren’t going away, we might as well take advantage of them. If you need help getting started, fire up your favorite search engine and ask it, “How do I start using AI?” You’ll probably get an AI-generated answer. And off you go…
Now What? Your Serve…
I am naively optimistic enough to believe there is value in trying to look at what I do, how I do it, and why I do it. Writing about things like those in this piece is part of that process. So to echo my ABCs of human motivation, the achievement of writing and sharing these posts delivers the benefit of helping me crystallize some thoughts and feelings, and tighten up some of the internal connections linking my body, mind, and spirit.
But this is not just about me. Sharing these thoughts and feelings also offers the promise of connection with others. That’s where you come in.
Your reactions eagerly sought and warmly welcomed. Feel free to leave your thoughts wherever you’re reading this if comments are supported, or to email me directly at medortch@dortchonit.com. I hope this begins some interesting conversational threads, excerpts and summaries of which I will gladly and gratefully share (anonymously or with attribution as you prefer) in future outings, several of which are already in various stages of construction.
Thanks for reading. Thanks again in advance if you share your reactions with me, share this post with others, or both. And please stay connected so we can continue and extend this emerging, evolving conversation.
We All Hunger and Thirst (WAHAT) is a series of pieces I’m writing and publishing to share things I’ve learned and foster connections and conversations about various elements of life and the world. Send your suggestions and reactions to me directly at medortch@dortchonit.com, and feel free to share what you’ve read with others.
By the way, I belatedly learned that “wahat” has multiple meanings that have nothing to do with me or my opinions. So I’ve added my name to what I’m calling this series of outings. Just to be clear about who is to blame. Thanks!
