Democrats Demand Answers on DOGE’s Use of AI

Democrats on the House Oversight Committee dismissed two dozen inquiries on Wednesday morning to obtain the federal authorities of the Federal Authority in order to get information about plans for the installation of AI software Federal authorities In the middle of the ongoing Reasons to the government Wing.

The flood of inquiries follows the recently carried out reports of WIRED And The Washington Post With regard to the efforts of Elon Musk’s so-called Ministry of Government Efficiency (Doge), to automate tasks with a variety of proprietary AI tools and access sensitive data.

“The American people trust the federal government sensitive personal information regarding their health, finance and other biographical information, whereby this information is not disclosed or not properly used without its consent”, in the inquiries “including the use of an unexpected and inexplicable AI software”. “

The inquiries received by WIRED are signed by Gerald Conlly, a Democratic Congress member from Virginia.

The central purpose of the inquiries is to demonstrate the agencies that potential use of AI is legal and that steps are taken to protect the private data of the Americans. The Democrats also want to know whether the use of KI Musk benefits financially, who founded Xai and its Difficult electric car companyTesla, works to turn towards robotics and AI. The Democrats are still concerned, says Connolly, that Musk could use his access to sensitive government data for personal enrichment and use the data to “complain about his own proprietary AI model, known as the GROK.

In the inquiries, Connolly states that the federal authorities are “bound to several legal requirements for the use of AI software”, which mainly indicates the program of the Federal Risic and Authorization Management, which can standardize and ensure that the AIS-based tools are properly assessed for security risks. He also refers to the progressive American Ai Act, the requires Federal authorities “preparation and maintenance of an inventory of applications for artificial intelligence of the agency” and “provide the public”.

Documents that were received by WIRED last week Used a proprietary chatbot Called GSAI at around 1,500 federal workers. The GSA oversees the federal government real estate and provides information technology services to many agencies.

A memo preserved by WIRED Reporter shows that employees were warned of all controlled information not classified information before feeding the software. Other agencies, including the departments for finance ministries as well as health and human services, have considered to use a chat bot, although not necessarily GSAI, according to the documents regarded by WIRED.

WIRED has also reported that the US Army is currently using software that is called Camogpt Scan his data estate systems For references to diversity, justice, inclusion and accessibility. A spokesman for the army confirmed the existence of the tool, but refused to provide further information about how the army wanted to use it.

In the inquiries, Connolly wrote that the Ministry of Education has personally identifiable information about more than 43 million people bound with programs for student aid. “Because of the opaque and frenetic pace in which Dog seems to be operating,” he writes, “I am deeply concerned that students, parents, spouses, family members and sensitive information from secret members of the Doge team for unclear purposes and without protective measures to prevent disclosure or inappropriate applications, unpathic applications, inadequate use, inadmissible applications. Washington Post previously reported This dog had started to feed sensitive federal data that comes from the Ministry of Education from data set systems in order to analyze their expenses.

The Minister of Education Linda McMahon said on Tuesday that she continued with plans to relieve more than a thousand workers in the department and to join Hundreds of others Who accepted Doge “Buyouts” last month. The educational department has lost almost half of its workforce – the first step, McMahon saysto completely abolish the agency.

“The use of AI for evaluating sensitive data is with serious dangers that go beyond inappropriate disclosure,” writes Connolly and warns that “the entries used and the parameters selected for the analysis can be incorrect. By designing the AI ​​software, employees can be misinterpreted to other concerns. “

He adds: “Without a clear purpose for the use of AI, guardrails, to ensure adequate remodeling of data as well as adequate supervision and transparency, the application of AI is dangerous and may violate the federal law.”