The United States, the United Kingdom, the European Union and several other countries have signed an AI safety treaty set by the Council of Europe (COE), an international standards and human rights organization. This landmark treaty, known as the Framework Convention on Artificial Intelligence and Human Rights, Democracy and the Rule of Law, opened for signature in Vilnius, Lithuania. It is the first legally binding international agreement aimed at ensuring that AI systems align with democratic values.
The treaty focuses on three main areas: protecting human rights (including privacy and preventing discrimination), safeguarding democracy and upholding the rule of law. It also provides a legal framework to cover the entire lifecycle of AI systems, fostering innovation and managing potential risks.
Apart from the US, the UK and the EU, other signatories to the treaty include Andorra, Georgia, Iceland, Norway, Moldova, San Marino and Israel. Notably, several major Asian and Middle Eastern countries, as well as Russia, are absent, but according to a statement from the Council of Europe, any country will be eligible to join in the future provided they commit to complying with its provisions.
“We must ensure that the rise of AI maintains our standards, not undermines them,” COE Secretary General Marija Pejcinovic Buric said in the statement. “The Framework Convention is designed to ensure just that. It is a strong and balanced text – one that is built on the open and inclusive approach it was drafted with and that ensured it benefited from many more expert views.
The treaty will enter into force three months after it has been ratified by five signatories, including at least three member states of the Council of Europe. The CoE’s treaty joins other recent efforts to regulate AI, including the UK’s AI Safety Summit, the G7-led Hiroshima AI process, and the UN’s AI resolution.
OpenAI will build and open source $500 billion worth of AI infrastructure in the United States over the next four years in partnership with SoftBank. The two entities have teamed up to set up a new company called the Stargate Project to build AI data centers for the ChatGPT maker, and according to their announcement, it will “secure US leadership in AI” as well as “create hundreds of thousands of US jobs.” SoftBank will fund the project, while OpenAI will be in charge of its operations. SoftBank CEO Masayoshi Son will serve as its chairman.
While OpenAI and SoftBank will serve as the Stargate Project’s lead partners, the initiative also involves a number of other companies. In addition to OpenAI, Arm, NVIDIA, Oracle and, of course, Microsoft will be its key initial technology partners. The company’s official announcement states that OpenAI, NVIDIA and Oracle will work together to “build and operate this computing system,” though it did not elaborate what that means.
As far as Microsoft is concerned, the company has posted an announcement about its involvement in the project and said it will continue its “strategic partnership” with OpenAI. Microsoft will still have the rights to use OpenAI IP on its products, and the OpenAI API will still run exclusively on Azure.
However, it will no longer be OpenAI’s exclusive provider of computing capacity. Microsoft will have a “right of first refusal” whenever OpenAI requires additional capacity as its needs grow, meaning the ChatGPT-maker will have to check with Microsoft before working with other parties.
The Stargate Project is immediately releasing $100 billion for its first data center buildout, starting with one in Texas. As TechCrunch notes, it previously reported that OpenAI was in talks with Oracle to lease a data center in Abilene, Texas, that could reach nearly one gigawatt of power by 2026. It didn’t mention any other sites, but it said it was evaluating “potential” locations across the country.