News & Insights
We speak with Manish Sood, CEO and Founder of Reltio, about artificial intelligence (AI) washing and steps companies can take to implement effective AI governance and compliance. Sood also shares how AI and automation are transforming data management.
This Week's Guest Spotlight
Manish Sood, CEO and Founder of Reltio
How widespread is AI washing in today's market, and what are the potential risks for companies and consumers?
AI washing is incredibly widespread in today’s market. Many companies are simply applying AI labels on their products to give the impression of being cutting-edge, with lofty claims about AI capabilities that are unlikely as advanced as they say, nor do they deliver business value or tangible outcomes. This can be extremely misleading for consumers who don’t understand the technology. There’s a learning curve needed for everyone, from enterprise IT to the average consumer, to understand what AI is and what it is not. Consumers interested in financial apps with AI capabilities, for example, will want to ensure they are getting what they bargained for.
AI washing poses significant risks for companies. There’s potential for regulatory fines and a risk of reputational damage in the marketplace. Government regulators are keenly aware of AI washing and have already taken action against some businesses for overselling their AI capabilities. The Federal Trade Commission and the U.S. Securities and Exchange Commission have both warned companies about making false claims about AI.
What practical steps can companies take to implement effective AI governance and compliance?
First, companies need trusted, unified data—in real time. Second, companies should be transparent about their data, ensure data quality, accurately present data-driven insights, and prioritize data privacy and security. Third, companies should ensure their customers have full control over their data and that the platform provides full transparency and audit on any changes to the data. Customers should always know what data is being used by a company’s AI-enabled capabilities.
Beyond that, consumers, researchers, and regulators should also critically evaluate AI claims and demand transparency to hold companies accountable for their AI assertions. Additionally, it’s important to provide clear information about data sources, processing methods, and AI model outcomes.
How are AI and automation transforming data management, and why is this a business imperative?
AI and automation are revolutionizing data management, significantly improving the time-to-value and enabling businesses to become data driven faster than ever. The introduction of LLM-powered, pre-trained machine learning features for entity resolution, for example, is transforming how businesses handle data unification. Reltio’s Flexible Entity Resolution Networks (FERN), for example, uses LLM-powered pre-trained models for rule-free matching with high accuracy. This innovative approach eliminates the need for extensive rule configuration, significantly reducing the effort and time required for data unification. The pre-trained models, incorporating zero-shot learning, suggest matches out-of-the-box with greater precision, boosting data team productivity and shortening implementation time.
There is also increased productivity for data stewards as they can resolve issues much faster and spend more time on value-added activities and less time “fixing data.” The Reltio Intelligent Assistant (RIA) for example, leverages generative AI and natural language technology, providing a natural language, chat-based interface integrated within the Reltio platform. This feature enables users to search digital content, including complex technical concepts, more efficiently. By delivering answers to complex technical questions quickly, RIA amplifies data stewards’ productivity, allowing them to focus on critical issues that improve data accuracy and drive the curation of high-quality data.
AI-powered data unification and management capabilities reduce the hundreds of person-hours previously required to define, configure, and iterate match rules. The automation of these tasks results in significant cost savings and allows businesses to allocate resources more strategically. With higher match accuracy and minimal effort required, companies can achieve better results with less manual intervention. AI and automation in data management enhance accuracy, simplify processes, boost productivity, reduce costs, and provide a strategic advantage. These benefits make the adoption of AI-driven data management solutions a business imperative, enabling companies to leverage their data more effectively and stay ahead in a competitive landscape.
Which specific trends in your industry are you most excited about?
The industry trend I'm most excited about is the intense enterprise focus on feeding AI with trusted, unified data to fuel their business initiatives. Without clean, connected, interoperable data, investments in AI are wasted. Diving into AI without a solid data foundation presents risks for failure. That’s why we’re seeing increasing interest in data unification and management solutions to deliver trusted data that can be used across the enterprise to unlock digital transformation and empower organizations to fully leverage the power of AI.
Do you have any unique predictions on the outlook of your industry?
Applications as we know them are going away as AI will open up the reinvention of the entire landscape. The current UI-based applications will soon be obsolete and replaced by AI-based conversational experiences. No one will need to go into an application to enter data or search for it. This can be done via a conversational AI experience where you ask a question using natural language. While conversational UX has been around for a while, we now have a usable technology (GenAI UX) to make it real. A natural language question can be used to update data, query data, outline a business intelligence report, and route a request, among other time-saving tasks. Conversational UX and agentic workflows are the future of applications, but they need data to power them - the data quality has to improve, and the data must be available at lower latency to ensure the best possible experience for users.
This interview originally appeared in our TradeTalks newsletter. Sign up here to access exclusive market analysis by a new industry expert each week. We also spotlight must-see TradeTalks videos from the past week.
TradeTalks Newsletter
Sign up to receive your weekly dose of trading news, trends and education. Delivered Wednesdays.
More Related Articles
This data feed is not available at this time.