Post by : Badri Ariffin

The acting chief of the U.S. Cybersecurity and Infrastructure Security Agency (CISA), Madhu Gottumukkala, finds himself under investigation due to reports of uploading sensitive government documents to the publicly accessible version of ChatGPT.

Media outlets have highlighted that these documents were labeled “For Official Use Only”. Although they are not classified, they remain sensitive and were not intended for public access. CISA's automated cybersecurity systems identified the uploads in August 2025, triggering an internal security alert.

Gottumukkala has been acting director of CISA since May 2025 and received temporary approval to utilize ChatGPT as part of a governmental initiative exploring artificial intelligence. However, the majority of Department of Homeland Security (DHS) employees had restrictions placed on their use of public AI platforms due to data security issues.

In response, the Department of Homeland Security has initiated a review to assess if the uploaded information jeopardized government security. The investigation's findings have not yet been disclosed.

This event raises alarms because public AI applications can retain user inputs and utilize them to enhance their systems. Conversely, DHS employs internal AI tools such as DHSChat, which are structured to protect government data within federal networks.

CISA has stated that Gottumukkala’s access to ChatGPT was restricted and short-term, with public AI platforms blocked for most personnel. This scenario has ignited a wider discussion regarding the safe implementation of artificial intelligence in government agencies while ensuring the protection of sensitive information.

#Global News