<span itemprop=”>
Organizations are looking to implement AI systems while safeguarding stakeholder privacy and maintaining trust.
At Data Summit 2026, Uchenna Okezie,senior analyst,O Enterprises, covered the critical intersection of responsible artificial intelligence (RAI) and privacy protection in organizational contexts during her session, “Responsible AI & Privacy: Trusted Data for Analytics at Speed.”
The annualData Summitconference returned to Boston, May 6-7, 2026, with pre-conference workshops on May 5.
“This study aimed to look at the organizational practices and governance mechanisms used to prevent privacy violations caused by AI systems,” Okezie said. “As the adoption of AI has surged so too have AI incidents.”
Common AI failures include algorithmic biases, explainability issues, environmental harm, cybersecurity breaches, job losses, and accountability problems, she explained.
There is a critical need for the implementation of responsible artificial intelligence to eliminate these risks.
Through years-long research, she explored evidence-based and actionable practices that enable organizations to implement AI systems while safeguarding stakeholder privacy and maintaining trust, along with providing practical solutions to address costly privacy violations, helping managers navigate responsible AI adoption with greater confidence.
The qualitative research conducted assessed the organization’s practices and governance in relation to AI.
The major findings of this study found that organizations implemented AI privacy governance through layered architectures. Privacy protections in AI systems require integration and embedding of technical safeguards. And a board should be formed that makes sure rules and regulations are followed ethically.
With the proliferation of AI, employees and personnel need to be continuously trained on the changing regulations and research regarding the technology, she noted.
“What could be true today, can be something completely different tomorrow,” Okezie said.
Susan DiFranco,enterprise account executive and former senior solutions engineer,Perforce Delphix, focused her part of the session on “Trusted Data for Analytics at Speed,” showing how teams deliver compliant, high-quality data wherever it’s needed across AI and analytics initiatives, including modern platforms like Snowflake, Databricks, and Azure/Fabric.
“As we are moving toward AI it’s important to start with your base,” DiFranco said. “And that base is your data.”
There is now a data access crisis, she explained. To get access to the data itself can take time but once you have the data, some of it may have sensitive information.
With Delphix, data scientists can get access to data within hours, cutting down the cost and barrier to entry. Masked copies of datasets are superior to synthetic data, she said. Masked data preserves machine learning outcomes. ML models predict patterns, not identities.
“We’re still giving you the information you need without exposing you,” DiFranco said.
Many Data Summit 2026 presentations are available for review athttps://www.dbta.com/datasummit/2026/presentations.aspx.