US Cyber Defense Agency Head Posted Sensitive Information Online

The interim chief of the country’s cyber defense agency reportedly uploaded sensitive government records into a publicly accessible artificial intelligence platform last summer.

Madhu Gottumukkala, who currently serves as interim director of the Cybersecurity and Infrastructure Security Agency (CISA), reportedly uploaded the material into a commercial version of ChatGPT despite the application being broadly restricted across Department of Homeland Security (DHS) systems at the time, according to Politico. The incident drew attention inside the department because Gottumukkala personally sought an exception to use the tool shortly after joining the agency in May, officials told Politico.

The records were not classified, but they included agency contracting files designated for official use only, a label applied to sensitive information meant to remain within government channels, according to officials who requested anonymity due to fear of retaliation. Automated monitoring tools inside CISA detected the uploads in early August and issued repeated alerts intended to catch unauthorized disclosures of federal data, the officials said.

Senior DHS leadership initiated an internal assessment to determine whether the exposure posed any risk to government operations. Officials said they were not informed of the review’s final determination.

CISA spokesperson Marci McCarthy said in a statement that Gottumukkala received approval to access ChatGPT under specific DHS safeguards and described the usage as temporary and limited in scope. She added that the agency continues to pursue artificial intelligence adoption consistent with President Donald Trump’s directive to accelerate U.S. leadership in AI development.