AI in Everyday Life

Have you asked ChatGPT anything today? 

If not, you’re falling behind the times. ChatGPT and other Large Language Model (LLM) chatbots such as Claude and Perplexity have improved so much since they first appeared on the scene a scant two years ago, that anyone who is not using them is not getting full value out of the Internet and is in danger of being left behind as a Luddite.

ChatGPT is already in the top 10 websites in use and because of its rapid growth it will be in the top 5 websites in the near future.

In our polarized society, people say and read many things that are based on conjecture or something they think they’ve heard, which may be distorted and untrue because of the broken telephone effect. Conjectures lead to rumor, then fake news and ultimately conspiracy theories. Try to add some intelligence to these conversations and avoid going off on these tangents. The best way to get some intelligence back into your life today is by using a chatbot.

In Canada, Pierre Poilievre, leader of the Conservative Party has advocated for the repeal of Canada’s carbon tax. “Axe the Tax” seems to have resonated with many Canadians concerned about the rising cost of living. But Poilievre never provided an estimate of how much money his policy would save the average taxpayer. 

It’s not that hard to find out. Just ask ChatGPT
→ If the carbon tax was removed, what would be the average change to Canadian household income?

Here’s the answer.

_____________________________

The exact change to taxpayers’ income from removing the carbon tax would depend on several factors—including where they live, how much fossil fuel energy they use, and whether the government replaces the lost revenue or cuts spending. But here’s a breakdown of what to expect based on available data:

1. Direct Savings from Fuel and Heating Costs

If the carbon tax were removed:
– Gasoline prices would drop by 14.3 cents per litre (current federal carbon tax rate as of 2025).
– Natural gas bills would go down, especially in provinces where it’s used for home heating.
– Households driving more or using more fossil fuels would see bigger dollar savings.

2. Loss or Reduction of the Carbon Rebate

Currently, the federal government returns most of the carbon tax revenue to households via the Canada Carbon Rebate (formerly Climate Action Incentive). In 2024 a family of four in Ontario received about $1,120 per year; in Alberta, about $1,800 per year.

👉 For most low and middle income families, the rebate exceeds what they pay in carbon taxes. So they’d actually lose income if the carbon tax and the rebate were eliminated.

3. Impact on Disposable Income

– High energy users would gain more from lower prices than they’d lose from rebates.
– Low energy users (especially urban renters, transit users) would lose money overall due to rebate loss.
– Top earners would benefit modestly from tax removal, as energy is a smaller portion of their expenses.

4. Average Canadian Household Annual Impact

Fuel/Heating Savings+$700 
Lost Carbon Rebate-$1200
Net Income Change–$500

👉 So for the average Canadian household, income would decrease by $500.

_____________________________

[ Note: If you ask this same question to ChatGPT you will probably get a somewhat different response. This is because Large Language Models are based on neural networks in which whether one node triggers an adjacent node to activate is determined by a probability. But the essence of the answer will be the same. ]

One area where AI has become extremely good is providing medical information. A major medical journal reported a study showing that ChatGPT gave more accurate medical diagnoses than doctors.
AI Chatbots Defeated Doctors at Diagnosing Illness

If you don’t have the time or inclination to read this full article, here is a brief summary that I asked ChatGPT to write:

A recent study published in the Journal of the American Medical Association explored the role of AI, specifically ChatGPT, in assisting physicians with diagnostic tasks. The study found that ChatGPT, when used independently, outperformed both groups of doctors—those with and without access to the AI tool—in diagnostic accuracy. However, when physicians used ChatGPT alongside conventional resources, their performance only slightly improved compared to those without AI assistance. This suggests that while AI has the potential to enhance diagnostic accuracy, its effectiveness is contingent upon how well medical professionals integrate and utilize these tools within their diagnostic processes.

A few months ago I wrote an article about Pneumonia Vaccines. I had seen a notice in a medical newsletter recommending the Prevnar 20 vaccination. I had gotten some pneumonia vaccinations 10 years ago but my doctor confirmed that I should get the Prevnar 20 booster. When I followed up doing some of my own research, I discovered that there was an even newer vaccine than Prevnar 20 that seemed to be better. 

ChatGPT was invaluable in this research. It’s important to note that the value of a chatbot comes from engaging it in a conversation, which is above and beyond just posing individual questions as you do in a Google search. The conversation progressed from general recommendations to specific recommendations for seniors and it was only when I probed about newer vaccines at the end that it provided information on CAPVAXIVE, the latest one. 

LLMs generally first respond with the information that has been most widely reported in their training data. You have to learn how to converse with AI in order to get to some of the more detailed information, as in this case. It’s worth reading through the transcript of this conversation to see these principles (the link opens a window at the bottom – scroll up to get to the beginning).
ChatGPT pneumonia vaccines research 

Many people think that when you have a respiratory infection, green mucus signals a bacterial infection whereas yellow indicates a viral infection. A Google search yields a variety of answers. What’s the truth here? 

Read this conversation with ChatGPT to get a good understanding of the significance of mucus color.

_____________________________

If you find that your friends have become boring, or are not interested in things that interest you, or are just not available to chat with you, make friends with an AI. Many AI chatbots have developed a warm supportive style of personal interaction. To many people, Claude has become a close confidant.

“Claude just feels … different. More creative and empathetic. Less gratingly robotic. Its outputs, they say, are like the responses a smart, attentive human would give and less like the generic prose generated by other chatbots.”

How Claude Became Tech Insiders’ Chatbot of Choice

If you prefer to skip reading the full article, here is a brief summary that I asked Claude to write:

This article explores how Claude, a chatbot created by Anthropic, has become the preferred AI assistant among tech insiders in San Francisco, despite being less well-known than ChatGPT. Claude’s fans, many of whom work in the AI industry, praise its emotional intelligence, creativity, and human-like responses. They use it for everything from legal advice to relationship counseling, with some considering it a close friend. Unlike previous versions that were seen as overly cautious, newer versions of Claude have undergone “character training” to develop traits like open-mindedness and curiosity.

_____________________________

I’m no longer an avid hockey fan but I did get caught up in the The Gr8 Chase, culminating on April 6, 2025 when Alex Ovechkin broke Wayne Gretzky’s record for the most career goals in hockey. I wanted to see the full history of the NHL players who had held the record for career goals. Google search results were all over the map so I used ChatGPT. 

One of the things this example shows is that chatbots are not restricted to passively retrieving data from their huge memory store that was created during their training but that they can also do calculations. In this case, Goals/Game was calculated to construct the table. 

After completing this table, ChatGPT asked if I wanted to see the results displayed on a graph. I had not thought of that but I said yes and received this really nice result.  This is not a graph that ChatGPT found on the internet. It was generated by some code that chat ChatGPT wrote specifically to use the data from the table.

There were a lot of articles in the sports media conjecturing which current NHL players might be able to one day match Ovechkin’s record. Here is the analysis I asked ChatGPT to do.

It’s worth looking through the details of this rather long chat. It demonstrates some important principles of how to use AI effectively. There were a number of places where I asked ChatGPT to add more information and to change its presentation of the tables and the graph. To use an AI tool effectively, you need to think of it as a collaborator and engage it to work together with you towards getting the best results.
NHL career goals

_____________________________

ChatGPT can also do pretty complicated calculations. Suppose you know what the monthly payments are on a mortgage but don’t know what the interest rate is. While there is a formula to calculate monthly payments given the interest rate, there is not a simple formula to do this reverse calculation. But ChatGPT can do it.

What’s the interest rate on a $400,000 mortgage amortized over 25 years if the monthly payments are $2,326?

You may not be interested in this math or understand it but it’s worth looking at the detailed ChatGPT result. This is very fine work with a clear step by step presentation and beautiful mathematical formatting, which is not easy to do.
Mortgage calculation

_____________________________

While there are a lot of separate AI models that specialize in “image generation”, ChatGPT can also now create pictures and illustrations very easily. It can be a lot of fun. Draw your own cartoons to post on your favorite social media sites.

→ Do a picture of penguins from the Heard and McDonald Islands demonstrating in front of the White House carrying protest signs.”

So there you have it. These examples demonstrate several ways to use AI effectively. Add some intelligence to your life, start chatting today.

[ For a more general explanation of AI, what it can do and how it works, see the earlier article Happy New Era ]

Author: Ernie Dainow

In university the emerging field of using computers to understand the brain by simulating learning and thinking captivated me, culminating in a Master’s degree in Artificial Intelligence in Computer Science. My interests evolved from doing research to building systems. I worked on large mainframe computers, personal computers and network systems. My expertise spanned software development for academic and scientific research, business and financial applications, data communications, computer hardware products and the Internet. After retiring I began writing, sharing insights and interesting discoveries that are not widely known or understood outside of the computer field. You can download my free books from Apple Books, Google Play Books or https://www.smashwords.com/profile/view/edainow

9 thoughts on “AI in Everyday Life”

  1. Thank you Ernie. This is a really well-written and engaging article. I certainly learned something about “chatting” with ChatGPT.

    1. I’m glad you found it informative. I’ve been following AI for 50 years and these chatbots have been very exciting advances, and we are only at the beginning of the new AI era.

  2. My experience hasn’t been as positive, although I haven’t tried that much. My beef is that many things, especially in politics, economics, etc. require a point of view to reach a conclusion. I don’t mean bias or ignoring facts, but some way of interpreting. I found the AI answers superficial and lacking in depth and sophistication. So worry that AI could dumb us down. Get us to accept a homogenized version when we need critical, penetrating thinking. I think AI is just a start and no subsitute for deeper research that goes after something unique, or unusual or important.

    1. I haven’t used AI very much for political analysis so you might be right. But if you’re not getting very good answers, then there are a number of things you can do. One is to clarify your question and provide more background and context for what you are asking about. There is a whole new field called “prompt engineering” which provides guidelines and even courses on how to interact with an AI. Another thing you can do is to respond critically to the AI and say something like “This response is superficial. Can you provide a response with more depth?” It’s important to probe an AI to get more information. An excellent example of this was my chat on Pneumonia vaccines. I didn’t get information on the latest vaccine until I asked more specific questions.

  3. Hi Ernie. I tried ChatGPT today. I submitted a picture of a bird that I took this morning and it enhanced the picture nicely. Upon further inspection it wasn’t the same bird. I called ChatGPT out on this and it called it a mistake. I tried three more times before I finally got it to say “I’m sorry”. This is unacceptable so I deleted the app from my phone. That was my first and last experience with ChatGPT. I hope you enjoyed the story.
    Ron

    1. Great story.

      I use ChatGPT image generation for fun but I don’t think it’s the greatest feature. ChatGPT is most useful for text generation.

      I think you may be acting a little hastily. If you were turning your first bowl on a lathe and it splintered, would you throw out the lathe? What happened to that old baseball spirit – three strikes before you’re out.

  4. Hi Ernie!
    Mostly, I agree with you. However, when “it” went out of its way to lie, then I draw the line. It’s not like it didn’t do a good job with the image generation, I could accept that. But it tried to deceive me and and it kept trying to deceive me. That’s more than 3 strikes… that’s “You’re thrown out of the game!!!”
    Here is the big issue… CEOs, and CFOs, and president’s all signed up to tell their developers to engage in in this deception… I’m not down with that!!!

  5. Yet another incredible post!
    I am a physician and I can easily see the day when a patient could tell their entire medical history in their own words and include all of the medical reports such as lab tests and X-rays and ChatGPT could create a differential diagnosis and order further tests and even prescribe medications if appropriate! Lots of legal and insurance issues, but essentially entirely doable. One of the complaints that one often hears is the loss of empathy, but I believe that ChatGPT has that one nailed as well! When I ask ChatGPT difficult or complex questions ChatGPT will often compliment me for being so astute and I fell feel lots of positive vibes running through me!!
    My chartered accountant is a really nice guy and he is absolutely brilliant and empathic (as are most of the CA’s I have dealt with over the years…a marked contrast with lawyers!
    Every year I meet with him and he explains in great detail my tax returns…and I can’t understand a word of what he is saying.
    This year I just took my tax return and entered it into ChatGPT without asking any questions and it interpreted my returns in a way that I could easily understand and even found a few areas that I needed to explore and it answered questions that I had!

Leave a Reply

Your email address will not be published. Required fields are marked *