The US Patent and Trademark Office Banned Staff From Using Generative AI

According to an inner guidance letter from April 2023 that was made available to WIRED through a public records request, the US Patent and Trademark Office prohibited the use of relational artificial intelligence for any goal last month, citing security concerns with the technologies as well as the desire of some tools exhibiting “bias, unpredictability, and malicious behavior.” The workplace is” committed to pursuing development within our agency,” according to Jamie Holcombe, the USPTO’s chief information officer, but they are also “working to deliver these capabilities to the office in a concerned way.”

Paul Fucito, press secretary for the USPTO, clarified to WIRED that people can use” state-of-the-art conceptual AI models” at work—but just inside the company’s internal screening environment. In an email, Fucito wrote that “innovators from across the USPTO are now using the AI Lab to better know the capabilities and limitations of relational AI and to create AI-powered solutions to pressing business wants.”

Outside of the tests environment, USPTO employees are barred from relying on AI initiatives like OpenAI’s ChatGPT or Anthropic’s Claude for job responsibilities. The use of any outcomes from the equipment, including AI-generated images and videos, is also prohibited by the assistance note from last year. However, employees of the Patent Office is look up registered patents and applications using some AI programs that have been approved by the company’s own people database. A$ 75 million contract was signed with Accenture Federal Services earlier this year to update its patent database with enhanced AI-powered search capabilities.

The US Patent and Trademark Office, an organization within the Department of Commerce, is in charge of protecting scientists, awarding inventions, and registering marks. It also “advises the president of the United States, the secretary of commerce, and US government agencies on intellectual property ( IP ) policy, protection, and enforcement”, according to the USPTO’s website.

At a Google-sponsored function in 2023, Holcombe, the creator of the direction letter, said federal government makes it difficult for the public field to use innovative technologies. ” Everything we do in the state is very terrible, when you compare it to the business world, right”? he said. Holcombe specifically cited bulky budgeting, procurement, and compliance techniques, arguing that they bag the government’s ability to quickly follow improvements like artificial intelligence.

At least for some purposes, the USPTO is not the only state agency that forbids employees from using conceptual AI. Earlier this year, the National Archives and Records Administration prohibited the use of ChatGPT on government-issued tablets, according to 404 Media. But soon afterward, the National Archives hosted an internal presentation that encouraged employees to” think of]Google’s ] Gemini as a co-worker”. Some archives officially expressed fears about the reliability of relational AI during the meeting. The National Archives plans to release a new open robot that uses Google-developed technology to access archival records next month.

Another US government authorities are using—or avoiding—generative AI in various methods. The National Aeronautics and Space Administration, for example, particularly banned the use of AI ai for sensitive information. NASA did consider, but, to experiment with the systems for writing script and summarizing research. The organization also disclosed last week that it is working with Microsoft to create an AI robot that you aggregate satellite data for easy searching. NASA scientists and researchers are the only ones who can use that application, but its purpose is to “democratize access to spaceborne data.”