New Sponsored Episodes house line Daily Podcast — Live from the 2024 HousingWire Artificial Intelligence Summit — With Editor-in-Chief Sarah Wheeler core logic Data experts Amy Gromovsky and Anand Singh.
Gromowski, the company’s director of data science, and Singh, vice president of GenAI propertiesights, discussed the importance of data in the development of artificial intelligence and provided advice for companies looking to integrate artificial intelligence into business processes.
The conversation began with a discussion of the role of data in AI development and the best ways for companies to package data sets. Singer explained that modern AI tools require more and more data than ever before. He went on to explain that when it comes to developing artificial intelligence tools, dataset size is only half the battle. Data accuracy plays an important role, especially when companies use artificial intelligence for critical decision-making processes such as mortgage underwriting.
Gromowski added that data sets must contain in-depth historical information to inform future AI-driven decisions. Data diversity—meaning data should come from multiple sources—also plays a huge role. Singh emphasized the importance of using accurate data for real estate decisions. CoreLogic uses data to create a source of truth for customers to make accurate decisions, he said.
Wheeler went on to raise the question of how to combine accurate data with larger data sets that AI models prefer. Open artificial intelligence use. Singh explained that a general-purpose AI model cannot answer detailed real estate-related tips without referencing general resources. Therefore, including accurate and diverse data is critical to developing artificial intelligence and collecting quality data.
“Your AI needs to be driven by data, but your data needs to be unlocked by AI, so it goes both ways,” Singer said.
The conversation has since turned to the responsible use of artificial intelligence. Gromowski breaks down responsible AI into three considerations:
- Data usage rights. CoreLogic ensures that its materials are used within appropriate legal contracts and obligations.
- ethics. The company avoids disparate impact of AI models to ensure that all AI-generated solutions are fair and equitable to consumers.
- risk factors. CoreLogic determines its ethical obligations and reputational risks before using or building AI solutions.
CoreLogic relies on internal governance structures to guide the use of AI solutions based on what is happening in the larger regulated market. Singh pointed to potential biases in the real estate industry when generating data and said CoreLogic’s internal AI governance team has a greater responsibility to evaluate front-end data.
Since then, Wheeler has dabbled in artificial intelligence cybersecurity. From a technical perspective, Singh believes that CoreLogic’s artificial intelligence model is not open to the public and is controlled by the company. CoreLogic also monitors artificial intelligence for security vulnerabilities and destructive malware. When it comes to data security, the company screens its data sets for anything that could harm the public.
Next, the conversation turned to how CoreLogic’s customers are leveraging artificial intelligence solutions in their business processes. Singer said the company has been using artificial intelligence for decades to assist clients with risk assessments, valuations, image capture and more. Today, artificial intelligence can assist in sales conversations, customer communications, software engineering, and other areas. Real estate agents can also use artificial intelligence to handle day-to-day tasks such as building listings, artwork, and other daily tasks.
The conversation concluded with a discussion of the diversity of CoreLogic’s data across multiple industries. Wheeler asked if data diversity would help CoreLogic better inform its customers. Singh said the diverse data set enables CoreLogic’s AI tools to generate a unique understanding of business processes across multiple areas of the industry and how those processes intersect in terms of value. Doing so ensures that any AI-generated solutions serve the best interests of all parties in a real estate transaction.
