The mortgage industry is buzzing about generative AI and Sergio Butkewitsch, vice president of data, is entering the chat providing his take on today’s trends and what’s coming next in the interview below.
Q: What is the difference between AI and Generative AI?
A: Traditional and generative AI are complementary, but the objectives of generative AI are more relative to synthesis – meaning, putting something together in a way that it captures the essence of many things. You can accomplish more sophisticated things with generative AI. Traditional AI is more about prediction – like a crystal ball. To put it into context, using interest rates as an example: you can leverage traditional AI to help predict if interest rates will go up or down but with generative AI, you’d be able to take that context a step further, adding a recommendation flavor to it. For example, because interest rates are X, I should apply this percentage of my business to home equity and this percentage to first mortgage. It is what they achieve, rather than how they work, that defines the difference between traditional and generative AI.
Q: What excites you most about AI and how it is becoming more commonplace in the industry?
A: From a business perspective, it removes instances where the end user must sacrifice one thing over the other – such as speed over accuracy. I am excited to deliver that to the borrower. From a technical perspective, what excites me is that generative AI is very keen to operate in the multimedia world (i.e., language, video, audio). Before generative AI, we would require longer development cycles and sophisticated tools, but generative AI is just a matter of faster computers and our immersion in digital experience. It is very exciting.
Q: What are some misconceptions about AI?
A: It is easy to misunderstand generative AI as a “kaboom” and after this explosion, you have a technological marvel. That is incorrect. It’s a journey – an evolution. We have been on this AI train for years, and it makes a difference to be an early rider on this journey. To that end, companies trying to get into AI now, with generative AI as their first foray into it, will require a lot of luck. ServiceLink does not need the luck factor as we have been here since the onset. Another misconception is that AI will replace humans in the workforce. Also, not true. AI does, however, remove the redundant, labor-intensive pieces of a process, freeing up time for humans to work on more complex problems and think more strategically, driving even more innovation.
Q: How is ServiceLink leveraging AI?
A: We are pioneers in the AI realm and have been leveraging AI within our EXOS® origination suite of products and services since 2015, but we always ensure that the technology is mature and vetted before applying to our processes. One such example of an AI-infused offering from ServiceLink is our EXOS® Title, which leverages AI, machine learning and cloud-based automation engines to generate an instant title commitment in a matter of minutes – not days. AI is not necessarily “new” – it goes back decades and is a legacy and heritage for us. This experience with AI, built over the years, positioned us well to continue the journey with generative AI. It was the natural next step for us. We are already using generative AI in the form of large language models. In general, they are models that can understand and predict the sequences of communication. Some of it can be extremely complex. That complexity requires a large model to tackle it.
Q: What are some AI trends in lending that you foresee being big in 2024?
A: In a nutshell: automation at scale will be the trend to watch in 2024 and will go beyond the barebones predictive AI that is somewhat mainstream today. For example, mass customization – which sounds like an oxymoron – will become more mainstream. To a borrower, that could translate into a very niche, individualized experience that is tailored to their needs due to large-scale automation. In addition, much faster resolution. A task that used to take a week of time can now be done in a couple of hours, with much greater accuracy. Imagine the combination of these factors on the borrower experience – tailormade to them, fast, accurate, exactly what they want, how they want it. With AI, that is possible.
For more information on EXOS, and how this technology is bringing greater automation, ease and transparency into the mortgage lending process, contact us today.