The company, which is the enterprise B2B software business of Capital One, said in a Monday (March 23) press release that these new capabilities solve the unique governance challenges posed by the volume, variety and lack of coherent structure of unstructured data.
Unstructured data such as documents and multimedia files accounts for between 70% and 90% of organizational data, according to the release.
The upcoming capabilities that will be added to Capital One Databolt will, through automated workflows, inspect unstructured data formats, identify personally identifiable information (PII) and protect the PII, the release said.
The solution will use tokenization and other data protection approaches that preserve semantic structure so that enterprises can make the data safe and usable for analytics and AI workflows, per the release.
“With these new innovations, Databolt will help organizations move past the security bottlenecks that act as a barrier to innovation,” Prashant Prahlad, senior vice president and head of product at Capital One Software, said in the release. “We are intent on establishing secure-by-default as the standard, giving enterprises the confidence to forge ahead with their AI initiatives without compromising their sensitive data.”
Capital One introduced Databolt in April, saying this tokenization solution helps businesses address data security challenges by protecting sensitive data at scale without sacrificing performance or speed.
The solution replaces sensitive data with secure tokens, reducing the risk that sensitive data is exposed during a breach. It also preserves the underlying data format so that companies can still run applications, manage third-party data sharing and safety adopt generative AI.
“Traditional data security methods often fall short with limited scalability and added complexity,” Capital One Software President Ravi Raghu said in an April press release. “Tokenization [provides] next-level security as companies navigate a growing need to protect their data and prepare for an AI-first world.”
The PYMNTS Intelligence report “Agentic AI Breaks Out of the Sandbox” found that as of November 2025, nearly 40% of chief product officers are open to granting AI agents meaningful autonomy.