CISA Director Under Scrutiny for ChatGPT Upload Incident
The head of the Cybersecurity and Infrastructure Security Agency (CISA), Madhu Gottumukkala, has come under scrutiny following the upload of “sensitive” contracting documents to a public version of ChatGPT, prompting an internal investigation, as reported by insider sources.
As the acting director during the administration of President Donald Trump, Gottumukkala allegedly shared these files with the AI platform last summer, according to multiple unnamed officials from the Department of Homeland Security (DHS) who spoke to Politico.
While the documents did not carry classified status, they were explicitly marked “for official use only,” indicating their sensitive nature and restricted distribution.
The unauthorized upload triggered automated alerts designed to prevent the compromise of government data, as noted by the outlet.
In August, senior officials conducted an internal review to assess whether any governmental infrastructure was compromised as a result of the upload. The findings of this review remain undisclosed.
Gottumukkala, who formerly held the position of chief information officer for South Dakota’s Bureau of Information and Technology, had secured special permission to utilize ChatGPT, a privilege not extended to most DHS employees.
One official remarked to Politico, “He essentially compelled CISA to permit his access to ChatGPT, and unfortunately, he misused it.”
Madhu Gottumukkala of CISA is under scrutiny for uploading sensitive materials to a public version of ChatGPT.
It is important to note that data entered into the public version of ChatGPT, developed by OpenAI, has the potential to be integrated into AI prompts, thereby exposing it to approximately one billion users globally.
In a statement addressing the incident, Marci McCarthy, CISA’s director of public affairs, appeared to mitigate the severity of the occurrence. She clarified that Gottumukkala had received authorization to use ChatGPT under strict DHS protocols and characterized his use as “short-term and limited.”
“Acting Director Dr. Madhu Gottumukkala last utilized ChatGPT in mid-July 2025, under an authorized temporary exception granted to select employees,” she further explained.
“CISA’s security measures remain stringent, blocking default access to ChatGPT unless specific exceptions are made.”
It was also emphasized that the agency, responsible for fortifying government networks against adversaries like China and Russia, is committed to furthering American leadership in AI, in line with an executive order issued by Trump in January 2025.
Responses from both CISA and OpenAI were not promptly available when reached for comments by The Independent.
Included in the uploaded materials were documents labeled ‘for official use only,’ signaling their sensitive nature.
Gottumukkala has been at the helm of CISA since May, when DHS Secretary Kristi Noem appointed him as deputy director. Throughout his brief tenure, he has been embroiled in numerous controversies.
Last year, an incident involving at least half a dozen CISA employees being placed on leave arose after Gottumukkala requested a polygraph test, which he subsequently failed, as reported by Politico.
He has since disputed this claim, asserting to a congressman last week that he does not “accept the premise of that characterization.”
The revelation regarding Gottumukkala’s ChatGPT usage aligns with a broader trend of AI adoption among U.S. workers, with twelve percent of adults indicating daily utilization of AI technologies in their occupational roles, according to a recent Gallup poll.
Source link: Aol.com.
