The potential of AVM and AI - opportunities and risks for appraisers

The use of automated valuation models (AVMs) in real estate appraisal is on the rise. Banks, for example, are increasingly using them in the financing process for standard properties such as apartments and single-family homes, while real estate investors frequently use them when making investment decisions.

This leads to concern among many appraisers that their profession may be endangered or obsolete in the future. AVMs are often used only in conjunction with an overall automated valuation of a property, which can be inaccurate, especially for commercial properties or specialty properties such as schools with less transactional data. AVMs are also criticized for their "black box" nature, as it is difficult to understand the factors that lead to a particular valuation.

However, there are ways to combine the strengths of AVMs and the human expertise of appraisers. AVMs can be used as a supporting tool for appraisers by extracting key insights and opening the black box to answer questions about how individual object properties are valued by AVMs.

A regression analysis was conducted to develop a model that uses AVMs to assist appraisers in determining inputs to appraisal procedures in a transparent manner, thereby improving efficiency and transparency and making appraisals more evidence-based.

This model allows the appraiser to review and adjust the AVM results as needed to improve the accuracy of the estimates.

Overall, the combination of AVMs and artificial intelligence offers the opportunity to improve the efficiency and accuracy of real estate appraisals without compromising the role of the human appraiser. It is important to view AVMs as a supporting tool rather than a threat to the profession and to integrate the benefits of AVMs and artificial intelligence into the appraisal process.