More on AI Use in Community Associations

AI is simply the new terminology for computer programming and automation.
Everything "AI" is doing now was done in the past.
It's simply faster and a more comprehensive now.

That said, a company that incorporates some automation can make it more efficient.
Cost effective is likely depended on long term gains vs. short term.

I think there is a place for AI, but you are right it has a long way to go when it comes to providing accurate information about niche or specialized topics.

I think it can be used to help owners find answers within their own documents, or generate service requests and those types of things.

With caveats, there is no question in my mind that AI can help enormously with more informed HOA decision-making and better HOA communications.

I disagree that the sort of demanding legal research (analyzing case law in a complicated case) of which you speak is relevant to HOA Boards.

I don’t know about AI, but simple logic would be a welcome addition in many HOAs. LOL.

In complete seriousness, this is why I recommend AI to HOA Boards and HOA managers.

How many times does this forum see a new member who does not think to even check the governing documents or state statutes?

This is exactly my point. ChatGPT makes things up - that's its function. You can't rely on it to give you accurate information. The law is exactly where my lawyer friend complained about inaccuracies. All of us long-time board members know that what you leave out may be more important than what you include when it comes to parsing legal documents.

What we really need are expert systems that can be relied on. But to the best of my knowledge we don't have any in area. I think we're still testing them in the medical arena - for example, in diagnostics. And when we do have them, they will be costly - unlike the stuff that's currently available for free. And to be fair, medical diagnostics is more complicated than simply finding legal citations since it involves prediction based on available knowledge. But even for the simpler expert systems geared toward the law, I can pretty much guarantee that few associations would be willing to pay the price for them. They'll go with cheap (and inaccurate) any day and twice on Sundays.

Finally, as I mentioned above, would you trust the results from the recently-announced Chinese AI product(s)? Why do you believe that the issues are any different for our AI products - other than we're somewhat aware of the Chinese government's biases when it comes to information that it allows its citizens to see? Do we understand our AI's biases, other than we know their creators' primary motive is making money? Does their creators' recent and obvious involvement in our national politics not make you think "hmmmm"? We do know that our AIs have identifiable biases when it comes to protected classes - there are products to identify such things. Do these things not concern you at all?

I suppose you could make an argument that ChatGPT who invents things is still better than a lot of the completely ignorant board members out there. I suppose I may agree with you in some low stakes instances. But not as a general rule.

People see "AI" and they trust it. ChatGPT is not Google on steroids - Google's own "use at your own risk" warning tells you that. The two NY lawyers who were sanctioned for making that mistake are telling you that. So does my own experience with AI on my laptop. I felt coerced. I felt it enough that it was waking me up at 3 AM saying "pay attention!" And once I removed the various products, this stopped. Bottom line: I trust my intuition with this stuff. I have over 30 years experience with information systems. And I'm sensitive to manipulation. If I'm noticing it to the point that it wakes me up in the middle of the night, there is something there that I need to be aware of.

And I'm totally starting a jihad (Dune reference).

Generative AI

Quote:

Generative artificial intelligence (Generative AI, GenAI,[1] or GAI) is a subfield of artificial intelligence that uses generative models to produce text, images, videos, or other forms of data.[2][3][4] These models learn the underlying patterns and structures of their training data and use them to produce new data[5][6] based on the input, which often comes in the form of natural language prompts.[7][8]

Y'all are proving my points here. Generative AIs like ChatGPT produce plausible-sounding new data that is consistent with the patterns and structures of what it has been fed. Note that it does not say that the new data is necessarily consistent with the information it has been fed, only with the patterns and structures of that information.

In other words, it makes up some amount of what it is spitting back out. Unless you are already very familiar with the subject area (like a lawyer), you won't spot the things that don't belong.

If well educated, experienced board members and managers - which I assume our regular posters are - don't understand this, what are the chances that the average Joe will understand?

Generative AI is not Google on steroids.

Recently I watched a video that showed how one large language model did math. It didn't use numbers at all. It used language to arrive at an approximation of numbers and mathematical functions. I didn't understand it all, but I was appalled by what I did understand - not least by the inefficiency of the process. And we wonder why AI is destroying the planet with its energy consumption demands. Math in one thing that the first computers did very well.

Yeah there's a lot I think you are not grokking here.

When newbies ask a question here, I will start posting Google AI answers (nothing they are from Google AI) and see how they compare to your answers and my answers.

LOL you should be doing this already regardless. Never ever have I met so many incompetent professionals than in HOA management land.

I love AI for replying to nasty emails from Karen Home owners.

writes nice professional emails devoid of my personal bias against the person.

I love it for making cool images for my HOA newsletter. You need an image of happy neighbors cutting a ribbon to a new dog park. AI will whip up something that would of taken my hours to create using photoshop.

I love auto meeting notes

I love making podcasts with google notebook LM

It's not so good at making newsletters yet, too verbose. Nor can it do a lot of other things, but it's a glass half full, not glass half empty tool.

In fairness to the managers, the job often doesn't pay much and they are subject to same abuse that board members come in for. More, in fact, because it's the manager's job to deal with these delightful people.

In addition, management companies often require their managers to handle more communities than they can in a reasonable-length work week. And that's without any emergencies that derail the manager's planned week. And then the boards on their other communities get upset because they're not getting what they need (and have paid for).

Sixty-hour weeks, abusive clients, ignorant boards - for not-great pay? Who on earth would want to have a job like that if you could make more money elsewhere? So the folks who are skilled enough to find other jobs do so.

It all boils down to economics. Boards (homeowners actually) don't want to pay enough for the management companies to hire top notch people, so we get whatever we can get and we make it work. Or not.

There is one management in my area that has bucked this trend. They pay their people well, because the company owners have realized that if they provide top-notch service, the clients will come. Not surprisingly the company has a waiting list ...

As I always say, you get what you pay for.

For some reason, many homeowners and boards think you ought to be able to manage a multi-million dollar corporation on a shoe string, so they often go for the low bid. This is what the low bid buys you. You can't make the demands if the workers don't think the job is worth putting up with guff.

Our roofers would never pull a stunt like that, but I believe in paying a fair price for good work. (We'll see how long it takes our owners to decide that I'm being ridiculous and replacing me with a board member who thinks you can volunteer your way to solvency.)

How to dumb down your research - use AI.

Bottom line: Using AI in its current state at this time is already better than not using it. In some cases much better.

AI is accelerating. Mistakes are being corrected.

Not using AI means you are being left behind.

Behind what?

This reminds me a lot of the early days of computerized business information systems. I can't tell you how many projects I was on where we tried to get the future end-users to tell us what they wanted their information system to do, and they had no idea. They just knew that they needed a system because everybody else had them.

If you can't answer the question "behind what", then you're in the same position as those users.

I may agree with you if I were in the process of, say, developing systems for use by association management companies and similar - in short, if I were in a competitive business where I was looking for every advantage I could find.

Association boards are not in a competitive business. Associations aren't either, although the developers and builders certainly are. I suppose it's possible that some buyers will go elsewhere because the board and the management company aren't using the latest and the greatest.

You're still going to be stuck with the fact that quality AI will cost money and far too many boards try to do things on a shoestring. I'm sure they'll try to use free stuff if possible. It will be interesting to see what the coming lawsuits over intellectual property and copyright will do the the eventual price of AI products - because settlement costs will be passed along to the consumer.

And to put it bluntly, the users of the current generation of AI products are basically unpaid testers who are working out the kinks in the system. The companies that produce these products LOVE you - just the way the owners of social media companies LOVE the unpaid content providers whose data they can monetize.

Back in the day, we used to refer to technological leaps as "the bleeding edge".

Google says "use at your own risk." I'd believe them if I were you - especially if you're using the product in an environment where liability can rear its ugly head. Also makes me wonder what AI is doing to the cost of cyber insurance. Users may not care, but insurance companies sure do.

Good question. Let’s break it down:

The statement "If you are not using AI, then you are being left behind" suggests that AI is creating a new standard of speed, efficiency, and capability.

Behind what?

Behind the people, businesses, and organizations that are using AI to work faster, smarter, and more creatively.

Behind the evolving economy, where AI tools are increasingly built into products, services, and workflows.

Behind innovation and competitiveness, since AI is helping users spot opportunities, automate tasks, and generate ideas at a pace that's tough to match manually.

Behind the learning curve, because the longer you delay, the more catching up you’ll need to do later when AI becomes even more embedded in everyday life and work.

In short: you're behind the future that's already unfolding.

Would you like me to also write a version that's a little sharper or more philosophical, depending on what you’re aiming for?

Guess what wrote the above?

I love it.

Good one.

I haven't read the entire thread (yet), but now that I'm back to posting, that means I'm back to providing links to useful articles that could help folks with various issues.

This one talks about solving three common problems AI use can create and how to resolve them - https://www.library.hbs.edu/working-knowledge/solving-three-common-ai-challenges-companies-face?utm_source=ActiveCampaign&utm_medium=email&utm_cAI%20challenges%20%7C%20Pitching%20products%20%7C%20Remote%20work&utm_campaign=WK%20Newsletter%204-23-2025

I'm glad CAI is sponsoring a webinar on AI and HOAs - the more you learn about the technology, the more equipped you should become in determining how you'd like to use it in your community, or if you want to use it at all. For me, my main concern is that HOA boards might use it to do all of its thinking for them, and that's never a good idea, especially if you don't have a clear understanding of how it works. From what I've seen AI learns from whatever people toss in it, so if you're throwing in nonsense, nonsense is what AI will give you, but you might not be able to tell - or want to, because you didn't want to do the work in the first place and thought this would be easier.

I didn't use any AI software. It was whatever Google used for its summary, about which it had nothing to say.

And while we're on the subject ego, I have no interest in what others are doing in their communities. I'm not harmed by it, I don't benefit from it (other than via my investments). I have far too much on my plate just dealing with my own community to worry about what others are doing except in theoretical terms.