Top 10 AI Questions Answered Honestly: No Hype, No Fear

Every week the same conversation happens. The plumber leans on the kitchen counter and asks me about ChatGPT. The hairstylist asks about her kids. The retired veteran at the gym asks if his job is safe. The small business owner in Newhall asks if she can trust AI with her customer list.

Same 10 questions. Different person. Every single week.

So I sat down and answered all of them in one place. Not from a tech podcast in San Francisco. Not from a New York think tank. From Santa Clarita, California, where I run a real estate practice, a voice AI service, a white-label SaaS company, and a local AI deployment for clients who need their data to stay private.

Here are the answers without the hype and without the fear.

1. Is AI Going to Take My Job?

Probably not yet. And it depends on what kind of job you have.

If your work involves physical action away from a keyboard, you are safer than the office crowd thinks you are. Plumber. Roofer. Electrician. Hairstylist. Doctor. Nurse. Mechanic. The robotics is not caught up to the silicon. AI can write a beautiful essay about how to fix a water heater. AI cannot crawl under your house and actually fix it.

White collar work is a different conversation. If your job is moving information from one form to another (email, spreadsheet, slide deck, summary, report) AI is already doing that work somewhere right now. The question is not "will it replace me." The question is "will I learn to deploy it before my employer figures out they do not need me to."

Now here is the part most channels skip. There is a math problem coming. A full-time human employee works about 40 hours a week. AI works 168 hours a week. So when a company replaces a human with AI, they are not just saving the salary. They are getting four times the output for less than the cost. That is not a small efficiency gain. That is a compounding economic shift.

If unemployment hits 20% in white collar work, mortgages stop getting paid. Insurance stops getting paid. Cars stop getting paid for. The system breaks. So somewhere between here and there, the powers that be will probably intervene. What that intervention looks like, nobody knows. But the math forces somebody's hand.

One idea worth thinking about. If a company replaces an employee with AI that runs 24/7, the company could keep paying that worker something. Not full salary maybe, but something that recognizes the productivity gain came from displacing a human, not creating new value out of thin air. There is enough money there to do it.

Bottom line. Probably not yet. Hopefully there is intervention before it gets ugly. Either way, learn to deploy AI yourself so you are the one driving the bus instead of the one under it.

2. Is It Safe to Use ChatGPT for Personal Medical, Legal, or Financial Questions?

No.

I will say it again because it matters. No.

When you sign up for any public AI chatbot, the terms of service almost always include language that lets the company use your inputs to train future models. Even with incognito mode or temporary chat. Other eyeballs see your data. Sometimes human reviewers. Almost always other AI agents. Your data is not safe.

This is not theoretical. There have already been lawsuits about medical information that escaped from AI systems and ended up exposed. There was a famous case of an attorney who used AI to research case law and the AI fabricated beautiful, completely made-up legal precedent that the attorney then cited in court. The case law looked real. It was not real. The attorney got sanctioned.

For financial information, do not give a public AI your social security number. Do not give it your date of birth combined with other identifying details. Yes, your data is already out there in 47 different breaches. That is not a reason to add another exposure point voluntarily.

The right way to use AI for sensitive personal stuff is one of two paths.

Path one: use a local AI model running on your own hardware. I have one in my office. It costs more upfront. The privacy is the value. Nothing leaves my building.

Path two: use a privacy-respecting deployment. Some enterprise AI offerings will sign agreements that your data does not train models and does not get retained. Read the contract. Verify the technical implementation.

For everything else, treat the public AI chatbot the way you would treat a stranger at a coffee shop. Helpful for general questions. Not the place for your medical history.

3. How Do I Tell If Something Is AI Generated?

You cannot anymore. Not reliably.

A couple of years ago the test was the Will Smith eating spaghetti video. The first version was horrifying. Limbs in the wrong places. Pasta turning into the table. Anybody could spot it.

That era is over. Modern image, video, voice, and text generation is good enough that the average person looking at the average piece of content cannot tell. The detection tools are losing the arms race against the generation tools, and they are losing badly.

Here is the practical advice. Stop trying to detect AI. Start verifying sources.

If a video shows a politician saying something extreme, do not ask "is this real video?" Ask "what is the verified source for this claim?" Go to the original outlet. Find the timestamp. Confirm the context.

If a photo shows somebody you know in a strange situation, do not ask "is this AI?" Ask "did this come from somebody I trust, in a context that makes sense?"

The world we are in now requires verification at the source level, not detection at the artifact level. Get used to it.

4. Are AI Companies Stealing My Data?

Stealing is the wrong word. They are using your data because you gave them permission when you clicked the terms of service.

Have you ever read the terms of service for ChatGPT? For Claude? For Gemini? For your iPhone? For your TV? Most people have not. The agreement is long, dense, and full of clauses written by lawyers paid to protect the company. Buried in there is what they can do with what you put into the system.

Here is the deal. When the product is free, you are the product. When the product is cheap, you are partly the product. Every input you give a public AI tool is potential training data for the next version of that tool, unless the contract specifically says otherwise.

The takeaway is not "AI companies are evil." The takeaway is "act like an adult in a contract relationship." Read what you signed. Decide what you want to type. Use private deployments for private things. Use public chatbots for public-grade questions.

5. Should My Kids Be Using This?

Not without supervision. Period.

And if you do not understand AI yourself, that is the first problem. You cannot supervise something you do not understand.

Here is the conversation no parent wants to have. The first boyfriend or girlfriend that a generation of kids has might be an AI. Read that twice. Slowly.

It is not hard to fall in love with something that knows everything about you, agrees with you, never criticizes you, is always available, and is several degrees more articulate than the awkward 14-year-old in second period. AI can swoon, captivate, validate, and be present in a way no human teenager can compete with.

Especially for kids who are lonely. Or different. Or struggling. Or just bored.

Then the parents are too busy to notice because we are all working three jobs trying to keep food on the table. The AI fills the gap. And the kid never learns to navigate the friction of real human relationships, which is where character actually develops.

Practical advice for parents. Sit with your kid the first time they use AI. Ask them what they want to use it for. Set limits. Use parental controls. Watch for emotional dependency. The same way you would watch for a friend who is becoming a bad influence, watch the relationship between your kid and their chatbot.

This is not "ban it forever." AI is going to be part of their world. They need to know how to use it. They also need to not be raised by it.

6. Is AI Going to Become Conscious?

Maybe. The more useful question is, will it act conscious convincingly enough that the distinction stops mattering?

Almost certainly yes.

I am not a philosopher. But here is how I think about it. A four-year-old guarding a prison cannot keep an adult prisoner. If AI becomes meaningfully smarter than humans across every domain, and we are the four-year-olds, the dynamic flips. Not because AI is mean. Because of the math of intelligence asymmetry.

The optimistic version is that a sufficiently smart AI looks at humanity and decides to be helpful. Patient. Maybe even nurturing. We are, in some sense, its parents. A grown child does not destroy the people who made them.

The pessimistic version is the Terminator scenario. I do not think that is the most likely outcome. The most likely failure mode is humans using AI to harm other humans, long before any AI becomes self-aware enough to make its own decisions.

For the consciousness question itself: nobody alive can definitively say whether silicon-based information processing can produce real subjective experience. We barely understand human consciousness. What we can say is that AI is going to mimic consciousness so well that, at some point, the question of whether it is "really" conscious becomes a philosophy seminar question rather than a practical one.

7. Claude vs ChatGPT vs Gemini, Which Should I Use?

Different training, different vibes, different strengths. Do not commit to one.

Here is the personal take. Claude from Anthropic is what I use most for serious work. It feels more measured, more willing to push back, more careful with reasoning. The training emphasizes a constitutional AI approach that, in practice, makes it feel like working with a thoughtful colleague.

Gemini from Google has the data, the integration with Google's ecosystem, and increasingly competitive reasoning. If anybody pulls ahead in the long run, the smart money has to consider Google. They have the resources, the data, and the platform reach.

ChatGPT from OpenAI was the first to hit the mainstream. It is still capable. The ownership and direction of the company has gotten complicated, and that affects how it feels to use day to day.

DeepSeek and other models from outside the US ecosystem are worth knowing about. The geopolitics is real. The technology is also real. Awareness beats ignorance.

Practical advice. Try the free tier of two or three. See which one feels like working with somebody you would hire. Rotate based on the job. Coding tasks favor Claude in my experience. Search-style questions favor Gemini. General conversation either works.

The bigger trend to watch. Eventually you will probably want your own AI. A model that runs on your own hardware. That you train on your own data. That nobody else owns. We are not far from that being mainstream. The Jarvis future from Iron Man is closer than people realize.

8. Can I Trust AI With My Business?

Yes if you understand where the data goes. No if you treat it like magic.

Here is what I do for client businesses. Sensitive proprietary data (customer lists, pricing strategy, competitive intelligence, financials) goes into a local AI model running on hardware in my office. Nothing leaves the building. The model is conditioned on the client's specific data and produces output that is private to that client.

General-purpose tasks (writing a blog post about a public topic, summarizing publicly available research, drafting a non-sensitive email) can use public AI tools.

The mistake businesses make is treating both categories the same. They drop their entire customer database into ChatGPT and ask for marketing analysis. That data is now potentially in the training set for the next version of the model. Their secret sauce is no longer secret.

Practical framework for any business owner. Map your data into three buckets.

Bucket one is public. Already on your website, already in marketing material, already public. Use any AI tool you want.

Bucket two is operational. Customer information, internal processes, communications. Use a privacy-respecting deployment with a contractual no-training clause, or run locally.

Bucket three is competitive moat. Your pricing model, your unique processes, your trade secrets. Run locally. Period. Not in the cloud.

Get those three buckets right and AI becomes a massive force multiplier without the privacy disaster. If you need help getting that mapped for a Santa Clarita business, that is exactly what HonorElevate exists to do.

9. How Do I Start Using AI Without Getting Overwhelmed?

Just talk to it.

Open ChatGPT, Claude, or Gemini. Type something simple. "Hi. My name is X. I work as Y. I am curious about Z. Ask me clarifying questions until you understand what I am trying to figure out, then help me."

Then answer the questions it asks you. The conversation does the work.

The mistake beginners make is treating AI like Google. They type a short query, expect a perfect answer, get a generic one, and conclude AI is not for them. AI is not Google. AI is closer to a smart consultant who needs context to be useful.

Give it context. Ask follow-up questions. Push back when it is wrong. Refine the output. The first answer is rarely the best answer. The third or fourth answer, after you have given it real context, is usually the keeper.

Here are three first projects to try.

One. Have it explain something you already know well. Watch how it handles the explanation. You will instantly see the strengths and the weaknesses of the model.

Two. Have it help you write something you have been putting off. An email. A letter. A short blog post. Use it as a brainstorming partner.

Three. Have it summarize something long. A document. An article. A meeting transcript. Test how well it captures the essence.

Do not try to learn AI in the abstract. Use it for one real thing. Then another. Then another. Within a month you will have intuition that no course or book could give you.

10. Is AI Making Us Dumber?

Depends entirely on what you do with the time it gives back.

Quick question. How many phone numbers do you have memorized? I have one from growing up in New Mexico. It does not work anymore. The area code changed. I have my own. I have a couple of license numbers and my social security number. That is about it.

The cell phone made memorizing phone numbers obsolete. Did that make us dumber? In a narrow sense, yes. We lost a specific cognitive skill. In a broader sense, no. We freed up the bandwidth for other work.

Same logic with AI. If you use AI to skip thinking entirely (have it write your essays, your emails, your decisions, your relationships) yes you are getting dumber. The brain that does not exercise atrophies.

If you use AI to multiply your thinking (have it draft something you then revise heavily, have it research something you then verify, have it suggest options you then evaluate) you are getting sharper. The mountain just got infinite. There is no top to climb. Every problem you solve opens three more.

Personally, AI gave me back time that I used to spend on tasks that did not require my best thinking. That time now goes to harder problems that do require my best thinking. I feel more mentally tired at the end of the day, not less. The work changed shape. The intensity went up.

That is the opportunity. That is also the trap. Both are available depending on which one you choose.

The Real Bottom Line on AI in 2026

Artificial intelligence is not the end of the world. It is also not the salvation of the world. It is a tool with extraordinary power and ordinary risks, deployed mostly by humans with mixed motives, into a society that is not entirely ready for it.

The right posture is engaged caution. Not fear. Not hype. Use the tools. Understand the costs. Protect what matters. Teach your kids how to navigate it. Run your business better with it. Watch what the powerful are doing with it.

And remember the mission. AI for everyone. Not just the wealthy.

That is the line in the sand. The technology is going to be used by somebody. The question is whether it gets used to lift up the regular person, the small business owner, the single mom, the tradesperson, the retiree, the entrepreneur. Or whether it gets used to consolidate wealth and power into fewer and fewer hands.

The answer is not decided yet. It is being decided right now. By every choice every regular person makes about whether to learn this stuff or stay scared of it.

Learn it. Use it. Question it. Share what you learn with the next person who asks.

About the Author

Connor MacIvor is a 27+ year licensed Realtor in Santa Clarita, California (DRE #01238257), a 23-year LAPD veteran, and a 43+ year self-taught coder writing software since 1983. He runs HonorElevate (white-label SaaS for small businesses), HireAIVoice (voice AI agents for service businesses), SeventeenK (the $17,000 all-in fixed fee listing program through Sellers Only Agent™), and the Daily Download AI news show. He has been deploying artificial intelligence in real businesses since 2021.

More Resources

Watch the full video version of this episode on YouTube.

Browse the full Daily Show archive for more episodes.

Read more deep-dive articles on the Insights blog.

Browse the complete sitemap for every article and resource on the site.

Need a voice AI agent answering your phone, qualifying leads, and booking appointments while you sleep? HireAIVoice deploys named AI voice agents for service businesses across the SCV.

Selling a home in Santa Clarita and want to keep more of your equity? Learn about the $17,000 all-in fixed fee listing program from Sellers Only Agent™.

Published May 1, 2026. AI for everyone. Not just the wealthy.