OpenAI made public its latest policy blueprint at an event hosted by the US Centre for Strategic and International Studies on Wednesday 13 November in Washington DC.
The blueprint details how the U.S. should maintain its leadership in artificial intelligence and how it will meet the huge energy needs of AI technology.
In the document, OpenAI also describes its vision for AI technology as a ‘fundamental technology like electricity’ that could provide significant jobs, GDP growth, and investment, and that ‘offers an opportunity for re-industrialisation that is not to be missed! ‘.
And, of course, it contains a gimmick that almost all Americans can’t resist: the revival of the American dream.
At Wednesday’s event, Chris Lehane, OpenAI’s vice president of global affairs, said they spent ‘a lot of time’ discussing AI infrastructure needs with the Biden administration and the Trump team.
When Trump takes office, he plans to repeal Biden’s executive order on AI, saying it ‘impedes AI innovation’; in addition, Trump has acknowledged that the US needs to expand its energy supply to remain competitive in AI, and has proposed easing licensing requirements for the use of fossil fuels and nuclear energy.
OpenAI has made it clear that it will work with the new Trump-led administration on policies regarding AI. Specifically, OpenAI’s policy blueprint proposes the creation of a ‘North American Artificial Intelligence Alliance’ and a ‘North American Artificial Intelligence Compact,’ which would aim to simplify access to talent, financing, and supply chains for AI technologies, as well as compete with similar initiatives by China on AI. Competition with similar Chinese initiatives in AI.
OpenAI proposes ‘Artificial Intelligence Economic Zones’ to promote cross-state collaboration
According to OpenAI, this AI collaboration would start with the US and its neighbours, and then expand to the US network of allies around the world, including countries in the Middle East such as the UAE.
To incentivise states to accelerate AI infrastructure permitting and approvals, the blueprint also suggests that US states and the federal government work together to create ‘AI economic zones’.
Lehane points out that as the U.S. enters the digital age, a lot of the economic gains are going to the coasts, and the relatively ‘backward’ Midwest and Southwest are becoming a potential core area for AI investment – places that have the land and capacity to build wind farms and solar arrays, as well as the potential to build wind farms and solar arrays. solar arrays, and potentially nuclear facilities.
Given that the U.S. Navy operates about 100 Small Modular Reactors (SMRs) to power Navy submarines, OpenAI proposes to use the Navy’s expertise to build more civilian small and medium-sized reactors as a way to increase nuclear energy capacity.
Lehane is also considering building a data centre in Kansas and Iowa, which ‘have a lot of agricultural data,’ to create a big language model or inference model based on agriculture. These facilities would not only serve the community, but would also make them ‘centres of artificial intelligence in agriculture’.
Nuclear and clean energy: supporting energy needs for AI development
A latest report states that ‘China will build as much nuclear power capacity in 10 years as the United States will build in 40 years’ and has been approving additional nuclear reactors in recent years. Citing estimates, Lehane said that by 2030, the U.S. will need 50 gigawatts of energy to support the needs of the AI industry and compete with China.
And, in the face of that competition, ‘we have no choice,’ Lehane said.
In addition, OpenAI expects to introduce a National Transmission Highway Act to expand electricity, fibre optic connectivity and natural gas pipeline construction, and further seeks new authority and funding to lift barriers to transmission planning, permitting and payments, as existing processes fail to keep up with AI-driven demand.
Notably, the European Union also yesterday released its first draft ‘code of conduct’ for general purpose artificial intelligence (GPAI) models, which sets out guidelines on risk management and provides a blueprint for companies to comply and avoid severe penalties.
