How I Use AI on MichaelRyanMoney.com
Yes, I use AI. A lot. Here’s what it helps with, what it doesn’t decide, and why I’m still responsible for what you read here.
By Michael Ryan · Last reviewed October 2026
Do I use AI to help create Michael Ryan Money?
Yeah. A LOT.
I’d be crazy not to.
I had a PalmPilot before the iPhone even existed. I’ve watched us go from looking things up in books, to searching Google, to carrying practically every piece of information in the world around in our pockets.
AI is another tool. A ridiculously powerful one.
But here’s the other half of that answer: I’d be crazy to trust it blindly, too.
I use AI to help me do the work. I don’t let it decide what’s true, what I believe, or what gets published under my name.
Why would I not use AI?
Some of us remember card catalogs. You wanted information, you went to the library, figured out which drawer to look through, found the book, found the shelf, and hoped what you needed hadn’t changed since the book was printed.
Then came the internet. Then Google. Then smartphones.
Nobody would tell me today, “Michael, if you really care about accuracy, don’t Google the IRS announcement from this morning. Go find a book at the library.”
That would be nuts.
New technology can make finding and working with information dramatically faster. That doesn’t make the information automatically right.
Google never eliminated the need to know whether the website you found was garbage. AI doesn’t eliminate that either.
So what do I actually use AI for?
Quite a bit.
Research. Comparing sources. Finding contradictions. Organizing information. Working through a question from different angles. Analyzing data. Drafting. Editing. Coding. Checking my work. And sometimes just asking it, “What am I missing here?”
AI helps me
- Find and organize information faster
- Compare documents and explanations
- Spot contradictions or missing questions
- Analyze data and calculations
- Draft, edit and explain complicated ideas
- Build and test tools, code and page components
- Challenge my assumptions before I publish
I still decide
- Which source actually controls the answer
- Whether a claim is supported strongly enough
- What matters in the real world
- What deserves emphasis or qualification
- What my experience tells me to question
- What I believe the reader needs to understand
- Whether something is ready to publish under my name
The goal isn’t to replace judgment. It’s to spend less time on the mechanical stuff so I can spend more time on the part where judgment actually matters.
AI is fast. That doesn’t make it right.
This is probably the most important thing to understand about AI.
It can be incredibly useful and still be wrong.
It can misunderstand a rule. Miss an exception. Rely on information that changed. Confuse two similar concepts. Or give you a beautifully written answer with all the confidence in the world… that happens to be wrong.
That’s why I don’t treat AI as the authority.
- Start with the actual source. IRS, Social Security, Medicare, SEC, Treasury, plan documents, statutes, official data or whatever authority controls the question.
- Use AI to help investigate. Compare documents, find the relevant sections, surface contradictions and identify questions worth checking.
- Check what actually supports the claim. A link existing is not the same thing as a source proving the sentence I wrote.
- Say when something isn’t certain. Sometimes the right answer really is “it depends” or “we don’t know yet.”
- I make the final call. The responsibility doesn’t move from me to the software because software helped.
What I won’t let AI make up
Some lines are pretty simple.
- A client story that never happened
- A personal experience I never had
- A quote somebody never said
- A source that doesn’t exist
- A statistic just because it sounds plausible
- A credential or accomplishment I don’t have
- My opinion just because an AI model thinks it sounds like something I might say
If I use a representative or composite example to explain a situation, it should be clear that’s what it is.
What I bring that AI can’t
This is the part that gets lost when the conversation turns into “human versus AI.”
I’ve spent nearly 30 years around financial planning and financial decisions.
AI can explain a Roth conversion. It hasn’t spent decades watching actual people make retirement decisions, misunderstand rules, panic in bad markets, put off paperwork, argue with a spouse about money, or discover that the technically correct answer worked a little differently in real life.
That experience changes the questions I ask.
Sometimes the important part isn’t, “Is this technically allowed?”
It’s, “Yeah… but what happens next?”
Or, “What would I check before somebody actually did this?”
Or, “What’s the one exception that could completely change this answer?”
AI can help me investigate those questions much faster. Experience helps me know which questions are worth asking in the first place.
The better these tools get, the more valuable good judgment becomes, not less.
When AI and the source disagree
The source wins.
And if the source isn’t clear, I’d rather tell you that than pretend the answer is more certain than it is.
That’s also why the financial claims on this site are supposed to point back to the authority that actually supports them rather than asking you to trust me, Google or an AI model simply because the answer sounds convincing.
You can read the more formal version of these standards in the Michael Ryan Money Editorial Policy.
What about your information?
AI creates a second, separate question from “Is the answer right?”
What happens to the information you give it?
If you’re using a tool, form or AI-powered feature on Michael Ryan Money, pay attention to what information it asks for and what that information is needed for.
Don’t put passwords, Social Security numbers or other highly sensitive information into a general-purpose AI conversation just because the box lets you type it.
The site’s handling of personal information is covered in the Privacy Policy. This page doesn’t replace that policy.
What if I get something wrong?
I fix it.
AI doesn’t change that responsibility either.
Financial rules change. Agencies issue new guidance. Numbers get updated. Sometimes I can interpret something incorrectly. Sometimes a source itself gets clarified later.
If you see something that looks wrong, outdated or just doesn’t make sense, tell me.
I’d much rather check it than defend a bad answer.
I’m not interested in pretending AI doesn’t exist.
And I’m definitely not interested in handing it the keys and walking away.
Use the tool. Know its limits. Keep a human responsible.
That’s how I use AI at Michael Ryan Money.