The White House is acting on growing concerns around artificial intelligence by promoting responsible AI use among government agencies and garnering participation from large tech companies in a public assessment of their generative AI systems.
The Biden administration said on Thursday it has a commitment from tech companies -- including Nvidia, which makes GPUs used in AI systems; ChatGPT developer OpenAI; and AI-chatbot developers Google and Microsoft -- to "participate in a public evaluation of AI systems," according to a statement. The evaluation will be conducted on a platform from AI development firm Scale AI at the AI Village at DEFCON 31, the annual hacking convention held in August in Las Vegas.
The evaluation will let thousands of AI experts assess how generative AI models align with best practices outlined in the White House's Blueprint for an AI Bill of Rights as well as the National Institute of Standards and Technology's AI Risk Management Framework. It will also provide information about the impact of generative AI models and how to fix issues, according to the Biden administration.
"AI is one of the most powerful technologies of our time, but in order to seize the opportunities it presents, we must first mitigate its risks," the White House said.
AI researchers and experts have issued warnings about risks the technology poses, ranging from the spread of misinformation to algorithmic bias and discrimination. AI pioneer Geoffrey Hinton recently left Google while voicing concerns about AI, and a group led by Elon Musk recently appealed to AI developers to step back and initiate a six-month pause on AI development.
Impact of the White House action
The initiative signals that the federal government is taking the rapid advances in AI and its impact on society seriously, said Alan Pelz-Sharpe, founder of market research firm Deep Analysis. Indeed, the White House also said Vice President Kamala Harris was meeting with the CEOs of Alphabet, Anthropic, Microsoft and OpenAI on Thursday to talk about ethical innovation in AI.
Despite the good intentions behind the action, Pelz-Sharpe said he's concerned that it may be "near impossible" for tech giants to follow the government's guidance -- particularly companies implementing generative AI tools.
"The reality is that generative AI is, for all intents and purposes, inexplicable to humans," Pelz-Sharpe said. "Nobody can tell us how it came to its decisions."
Alan Pelz-SharpeFounder, Deep Analysis
Additionally, Pelz-Sharpe said tech companies have already made significant investments in the technology, which means there will be little to stop it from advancing.
"My hope is initiatives such as these will at least spark a more informed discussion involving both the public and government," he said. "Generative AI does represent a technical breakthrough. But even those that built the systems admit the breakthroughs came faster than they expected, and little thought has been given to its true impact."
Along with the public evaluation of tech companies' generative AI models, the White House also announced $140 million in funding for the National Science Foundation to launch seven new national AI research institutes.
The Office of Management and Budget is also announcing the planned release of draft policy guidance for government use of AI systems, which will be available for public comment this summer.
Makenzie Holland is a news writer covering big tech and federal regulation. Prior to joining TechTarget, she was a general reporter for the Wilmington StarNews and a crime and education reporter at the Wabash Plain Dealer.