AI can write the paragraph. It cannot sign the certification - by Josephine Aberle

AI helped me write this article.

I know that sentence bothers some people. It may sound like a confession, a shortcut, or proof that another professional skill is quietly being handed over to a machine. But that is not how I see it.

After more than 20 years in commercial real estate valuation, I know that the hardest part of the work is not finding data. It is knowing which data matters, which assumptions deserve caution, and which conclusions can be relied on when real money is at risk.

In that context, AI is not replacing judgment. If anything, it is forcing me to define judgment more clearly.

That matters in today’s commercial real estate market. We are not operating in a simple cycle in which yesterday’s comparable sale, last year’s rent growth, or a familiar capitalization rate can be accepted without question. Capital markets are still adjusting. Property sectors are recovering unevenly. Office assets, multifamily developments, affordable housing, industrial properties, data centers, and special-purpose assets each carry different risks. In this environment, faster analysis is helpful. Unsupported confidence is dangerous.

So, the better question is not whether AI helped write this. The better question is whether a qualified human being knew enough to challenge it.

Real estate has never been short on information. Appraisers, lenders, brokers, investors, developers, and asset managers are surrounded by sale comps, rent comps, expense benchmarks, cap rate surveys, construction budgets, operating statements, offering memoranda, absorption studies and financing assumptions. AI gives us more words, more models, more summaries, and more speed.

But in uncertain markets, what real estate often lacks is not data. It is trust and judgment.

We have an abundance of inputs, but the problem is that those inputs do not interpret themselves.

That has always been true in appraisal. A sale is not automatically comparable because it closed. A cap rate is not supportable because it appears in a survey. A conclusion is not reliable because the report is long.

AI cannot replace the professional accountability that makes a value conclusion credible. AI can generate language, but it cannot assume professional responsibility. In appraisal, responsibility is the product. 

That responsibility is not theoretical.

AI may help identify issues, but it cannot determine their significance in the context of a specific property, market, client, and intended use. It does not understand reliance. It does not defend a credit decision, answer to a review committee, testify, or sign a certification. It can produce a technically complete paragraph without understanding whether the underlying analysis is sound.

The appraiser’s signature is not decoration. It says a qualified professional considered the relevant data, applied appropriate methods, developed credible analyses, and is willing to stand behind the conclusions. It also says the appraiser remained in control of the analysis.

The professional must verify the facts, test the assumptions, reconcile the evidence, and decide whether the conclusion is supportable. 

Can the client rely on this value? Are the assumptions supported? Are the risks clear? Has the appraiser explained the reasoning, or merely stated the conclusion?

I see AI less as a threat to valuation professionals and more as a test of our professional discipline. My work was never just typing, summarizing, or formatting. My value is in judgment, credibility, and recognizing when a conclusion is not adequately supported.

There is also a personal side to this.

Like many professionals in real estate, I have had to rethink what my experience means in a changing market. The industry is different than it was five years ago. Hiring is different. Client expectations are different. Technology is different. The old assumptions about career paths, production models, and professional value are being tested. That kind of transition can be uncomfortable, but it can also be revealing. Using AI does not make me feel less experienced. Instead, it reminds me why experience matters. One person may use AI to generate a paragraph and assume the work is complete. Another will read the same paragraph and ask: Is that actually true? Is it supported? What is missing? What would a lender ask? What would a regulator question? What am I prepared to sign?

That is not resistance to technology. That is professional discipline.

The real risk is not that AI will be used in real estate. It already is. The real risk is that it will be used without the expertise needed to evaluate what it produces.

For appraisers, reviewers, lenders, and real estate decision-makers, the future of real estate valuation will not be defined by whether AI is used. It will be defined by whether qualified professionals remain accountable for the analysis, the reasoning, and the conclusions others rely on.

So yes, AI helped me write this article. It helped me organize my thoughts, test my wording, and sharpen the argument.

But the perspective is mine. The experience is mine. The judgment is mine.

And in appraisal, that part still matters most.

Josephine Aberle, MAI, operates Precision Valuation Review, LLC, Glastonbury, Conn.