Beyond Search: Practical Ways to Incorporate AI into Your Everyday Work and Life

Do you want to use AI more, but don’t know where to start?

“I know I should be using AI more, but…”

“I know I should be using AI more, but…” If this sounds like you, then read on!

So many people I’ve spoken with are either not using AI at all (one person referred to themselves as “Pre-AI”) or are only using it in a limited capacity (“I use it to help me write emails”).

My hunch is that at least part of this boils down to simply not knowing what to use it for (the other part is that people have told me, “I don’t need AI, I just use Google”)… which kills me because (not that I don’t love search, but) AI is so much more than search! (Although, ironically ChatGPT’s homepage now very closely resembles a search home page, and Perplexity.ai has really taken off with their search capabilities. But treating AI like an alternative to your favorite search engine means that you’re leaving a lot of value on the table.

This article aims to help you use AI more fully and give you some inspiration by sharing real, specific and practical examples of some of the ways that we’ve been using AI in our day to day lives.

Before we get in to the examples, I should also note that these will work with many of the Large Language Models available today. To learn more about many of the Large Language Models available today, you can download our free guide.

Some of the benefits of using AI for daily tasks include: Time Savings, Up-Skilling, Organization, Writing, Formatting, Coding and more!

1. Up-Skill & Save Time: Ask ‘how to’ questions for using software rather than reading through the help documentation. I recently started working with a whole new set of software, including AirTable, Softr, Beehiiv, and more. When I have a question or am not sure how to do something with said software, I now have a new go-to process. Before I go to the software’s help center and click around reading through the documentation, I ask ChatGPT (or my LLM of choice) my question. Often, it knows the answer and breaks it down for me step by step. Pro Tip: If it gives an answer but you need it broken down further, you can follow on with “Thanks! I’m a beginner and I need this information broken down step by step.” Boom.
2. Data Entry Be Gone! My kids’ school recently released the 2024–2025 school year calendar (the one that lays out all the days off and early dismissals starting this fall). Normally I take that calendar and painstakingly turn it into about 50 Google Calendar invites for me and my husband so that we can keep track of everything from Parent-Teacher conferences to Spring Break. But that takes so much time! And also, every single parent in the school has to do it individually for themselves. So this year, I did something differently. I asked ChatGPT’s vision capability to “read” the calendar document and turn it into a CSV file that I could then upload directly to my Google Calendar. I told a couple other mom’s about it and they asked for it too. It saved each of them at least 30–60 minutes of mind-numbing data entry!
3. Summarizing and Organizing Info: Summarizing your own writing to re-use for various content: Since I create so much content these days, and sometimes it needs to be long-form, and sometimes it needs to be short-form, I can take my finished article that I’ve written and ask the LLM to do all sorts of things: Suggest edits, Suggest a title, Summarize 3–5 key take-aways, Create a LinkedIn Post about it, and on and on. That saves me a bunch of time and keeps me organized!
4. Soften the blow: Need help with a soft-landing on messaging something tough or awkward? This one isn’t my use case, but one of my friends recently texted me to show me a text-message that ChatGPT created for her, helping her word-smith a message about a sensitive topic in a way that was kind but firm. These types of things aren’t everyone’s strong suit, and they aren’t fun to write, so they are easily put off and procrastinated. With a trusty LLM by your side, you can get help and support to make an uncomfortable task a little easier on everyone.
5. Coding: AI is a great coding companion! And there are coding-specific tools (like GitHub Copilot and Devin) that aim to specialize, but even working with a general LLM like Llama or ChatGPT can provide a ton of value, especially if you already have some basic tech skills. A couple years ago I took a Python course and so I am fairly comfortable around creating Python apps, but the speed with which I can generate scripts now has allowed me to be orders of magnitude more productive. I have also found that if I’m not exactly sure where to start, I can first share with the LLM the problem that I’m trying to solve. By giving it more context, it creates a more useful set of code for me to implement.
6. Brainstorming: Brainstorming is one of my favorite uses of an LLM! It’s so helpful to get out of your own head and bounce ideas back and forth. And if you are not sure where to start, you can even say “I’m not sure where to start, but I have this task…” and see how the conversation unfolds from there. I recently used this brainstorming capability to see what it would look like to create a 4-week content plan — something that I’ve never done end-to-end on my own. The LLM helped by organizing my ideas, and providing some ideas and suggestions of its own. I saved a lot of time and ended up with a better outcome than if we were working alone. Here are some other examples:

“What questions should I ask about xyz objective?”

“This is my idea but I need to refine it. Can you help?”

“What do you think about xyz?”

Pro-Tip: Make the LLM more powerful by giving it a persona. For example, in the case above I needed someone who knew a lot about content marketing, so I started by saying, “Imagine that you are an experienced and talented content marketer…” That helps the LLM focus it’s efforts most suitably for your conversation. This tip is part of a larger set of “Prompt Engineering” tips that are table stakes for getting the most of out using LLMs.

7. Read a menu! This is another one of my favorite use cases, and one that I wrote about in our inaugural Newsletter. I was in London a couple weeks ago, having dinner with an old friend at a very nice little restaurant (highly recommend). In the decade-plus since I’ve lived in the UK my main restaurant-going criteria has shifted from charm and quality to “do they have a play area.” So, needless to say, my Michelin-Vocabulary-Decoder is a bit rusty. And even though the waitstaff was exceptionally knowledgeable, I wanted to see if ChatGPT could help me out. So I took a picture of the menu, gave ChatGPT a basic prompt around helping me understand what was on the menu, and hit go. Not only did it come back right away with explanations of some of the more fancily-worded dishes, it also cleverly skipped over the easy stuff (bread & butter, olives). I was so impressed, and quite delighted!

8. Identify a geographic location. This is a fun one! My father-in-law was recently on vacation in the Mediterranean and he was hopping around various cities and islands. He sent over a picture from one port stop and there wasn’t much context except for a rather large sign that said something in another language. I used the “vision” capabilities of the LLM — which you can get for free from some services, like Google’s Gemini, to ask where the picture was taken. It returned the name of the city. I cross-checked it with Google Maps and then made it known to my FIL that I knew where he was. I earned some brownie points and learned something new about a new place in the world. Win-win!

I hope this inspires you to use AI more and find ways to save time!

OK so these were just a few examples to get your juices flowing. I hope this inspires you to use AI more and find ways to save time, learn new skills, and get better outcomes in whatever it is that you need to do going forward. If you have questions, feel free to reach out any time!

Please keep in mind the important considerations below!

Before we wrap though, I would be remiss if I didn’t also point out some really important considerations that everyone should keep in mind when using AI:

• Hallucinations — AI can produce hallucinations. This means that the AI creates information that is false or inaccurate and does so with a lot of confidence. It’s tough to spot a hallucination because usually it sounds plausible and reasonable. Therefore it’s important to always proof-read, cross-check, and edit anything produced by AI.
• Privacy — Many LLM companies will train their models from the input that you provide. Because of this it is important to never input any sensitive or private information. Additionally, if you want to turn off the training capability, some of the models have that feature. In ChatGPT, you can turn if off by Clicking on your name on the bottom left corner >> Settings >> Data Controls >> and then changing the setting under “Improve the model for everyone” where you can toggle On-Off

• Smarts — LLM’s are trained on enormous corpus of data and information. Therefore LLM’s may have more context and background knowledge on the topic you want to discuss than you might originally expect. Test it out by asking it what it knows about xyz.
• Validate! — I said this above with regards to Hallucinations, but this is not to be taken lightly and deserves repeating — Always proof-read and edit the output from an LLM and always validate the accuracy of what it says.
• Manners — Many people I’ve spoken with (myself included) find ourselves thanking the LLM after it provides some really helpful feedback. Now there are studies being done to determine what is the optimal level of manners to use in order to get the best performance out of an LLM. Check out this study from earlier in 2024 for more. They say, “We observed that impolite prompts often result in poor performance, but overly polite language does not guarantee better outcomes.”
• Prompt Engineering — This deserves it’s own article, and there are lots of guides out there to help you with this. The main thing to know is that the way you craft your inputs is highly relevant to the quality of output that you get. The other thing to keep in mind is that you can go back and forth in conversation with an LLM to get to your desired output. You don’t have to stop with just one prompt. Tips like giving the LLM a persona and explaining the format of the output that you want will go a long way!