AI Agents and Appraiser Oversupply

[CGinMN]

In programming, you will get better programs and more programmers. (To a degree, we will all become programmers as Ale is demonstrating.) In appraisal, you will get better appraisals but fewer appraisers. The difference is that real estate is finite, whereas software is infinite. In both cases, AI changes the professional's role and relationship to what they do, but it doesn't eliminate the need for the human.

 

[Vermonter]

Because it does not simultaneously synthesize information and has no moral guideline or end goal. The books I am reading say they do not know if AI will ever be able to do that - and if it can it is a decade or more out -

AI can be taught all that very easily. Today, not a decade or more out.

What do you mean by Better?

More accurate valuations. In a highly conforming market, AI can produce more accurate results right now. The limitation is the data it's fed. Feed the right model, the right data with the right parameters and it can provide an appraisal report in minutes with more support than a human appraiser could in a week.

And If AI turns out to be "better" at most things than humans are, what will happen to all the humans whose jobs and income disappear?

Literally what I'm trying to tell everyone. No one is going to slow AI progress down to preserve jobs or so you can find a new profession. All the players are in a race and Deepseek open sourcing things just put it all out to the world for anyone with a computer to use.

The genie is out of the bottle

 

[Vermonter]

As to your second part, AI can't differentiate quality of construction, slipshod DIY remodeling, end of useful life Roofing, deferred maintenance, Etc....yet. This is where the human appraiser shines.

This may be where a human shines, but you don't need an appraiser for those things

 

[Carnivore]

I ask if AI if it could analyze a house interior from pictures

Yes, AIs can indeed analyze photos of residential house interiors to assess their condition and quality. By leveraging computer vision and machine learning algorithms, AI can identify various aspects such as:

- **Cleanliness and Maintenance:** Detecting dust, dirt, and wear and tear.
- **Damage:** Identifying cracks, water damage, mold, and other types of damage.
- **Furnishings and Decor:** Evaluating the quality and arrangement of furniture and decor.
- **Structural Aspects:** Checking the condition of walls, floors, ceilings, and other structural elements.

These capabilities are widely used in industries like real estate, insurance, and property management to streamline inspections and valuations. How do you feel about this use of technology in everyday life?

 

[George Hatch]

And If AI turns out to be "better" at most things than humans are, what will happen to all the humans whose jobs and income disappear?

The individual's choice is to either join in or don't. The adoption and integration of AI hinges on the actions at the societal level, not the individual level. If you or I won't do it then there are plenty of others who will because it will eventually pay to join in and it will cost to not join in. Theoretically, anyway.

Appraisers in particular cannot believe in the principle of substitution and still engage with the idea that the market participants won't go with the cheaper alternative, if/when available. As for those who sit it out they might not be the only ones who eventually get passed up by the market but they will be among the first to lose. In theory.

There doesn't exist a "just say no" choice that our society or any of our competitors can make. Technology doesn't sleep.

 

[Terrel L. Shields]

Floorplans and digital walkthroughs will be created by homeowners via smartphone

which suggests some pee-poor house plans. HO's are not exactly excellent architects and AI isn't going to make such judgments for them.

Devoting so much resources for a small output of improvement

Our fees preclude integrating AI into our work flow outside simple peripheral issues. That's the problem now with 'showing your work' in the adjustment category. Spend too much time actually doing those magic paired sales, or running an exhaustive multi-linear regression and you've got 4 or 5 hours in a report before you even start.

when AI starts taking away all this employment for humans, what are humans going to do?

riot and destroy the computers... seriously. They are not going to meekly starve to death nor are they willing to set at home and twiddle their thumbs while drawing a check to live. Humans are not built that way.

And another factor. N. N. Taleb predicted the failure of FNMA and why our financial system was a house of cards that would topple over as domino hit domino. The same problem is AI's problem. What happens when AI has a self-created glitch. Or, it simply takes off on its own course of action without regard to its original purpose. A few years ago, an AI experiment had to be shut down because the computers had created their own language the engineers couldn't understand.

This incident primarily refers to an experiment by Facebook's Artificial Intelligence Research (FAIR) team where two chatbots, nicknamed "Alice" and "Bob", were designed to negotiate with each other, leading them to develop their own cryptic language.​

Why they shut it down:​

While the AI creating its own language might seem alarming, the researchers shut down the experiment not because they feared the AI, but because the bots were no longer communicating in a way that was understandable or aligned with the intended goal of the study.​

 

[J Grant]

In programming, you will get better programs and more programmers. (To a degree, we will all become programmers as Ale is demonstrating.) In appraisal, you will get better appraisals but fewer appraisers. The difference is that real estate is finite, whereas software is infinite. In both cases, AI changes the professional's role and relationship to what they do, but it doesn't eliminate the need for the human.

Not everyone can retrain to be a programmer - what do you think everyone else, the vast majority of people in an array of professions, trades, and work, will do?

What is a "better" appraisal? A MV opinion is never provable, it is a model.

And how will "better" appraisals make more money for lenders when mortgage volume is a limited product and volume contracts and expands with the market?

AI will not eliminate the need for humans but the marginal degree of "better": It provides might come at a steep human cost -

 

[CGinMN]

As to your second part, AI can't differentiate quality of construction, slipshod DIY remodeling, end of useful life Roofing, deferred maintenance, Etc....yet. This is where the human appraiser shines.

This may be where a human shines, but you don't need an appraiser for those things

Whatever allows the human to identify this information in photos, it can be taught to and learned by an AI analyzing those same photos. One incentive to keep you out of houses is to level the playing field. AI can't do smells (cat ****) or proprioception (squishy floor felt with your foot), that is until we have robots.

What is a "better" appraisal? A MV opinion is never provable, it is a model.

A good example of how to define what a "better" appraisal is would be to think about your appraisal reports, then compare it to others you have seen. Pretty obvious, no?

 

[George Hatch]

Developing the skills to use an app won't make someone a programmer. They'll just be another user, same as people who are capable of working in Excel or Powerpoint.

 

[J Grant]

The individual's choice is to either join in or don't. The adoption and integration of AI hinges on the actions at the societal level, not the individual level. If you or I won't do it then there are plenty of others who will because it will eventually pay to join in and it will cost to not join in. Theoretically, anyway.

Appraisers in particular cannot believe in the principle of substitution and still engage with the idea that the market participants won't go with the cheaper alternative, if/when available. As for those who sit it out they might not be the only ones who eventually get passed up by the market but they will be among the first to lose. In theory.

There doesn't exist a "just say no" choice that our society or any of our competitors can make. Technology doesn't sleep.

Workers and professionals can adapt and learn whatever AI we need to - most of it will run in the bg or integrate with software - we are not programmers or coders, and there is no need to learn that, nor could we complete it if we did.
Some displaced workers can be returned, others can not or will not and there may not be enough jobs even for the ones who do.

But when AI in other fields, or even our own field, eliminates a large mass of jobs, what is your proposed solution? National leaders who set policies, company owners and business owners who decide to use human labor instead of an AI alternative might be one solution, A UI to replace some lost income another, go back to a simple barter economy live a very low-cost lifestyle for others, others will go postal and on ramps and others will turn to criminal activity. That is what I see as some outcomes -