On a rainy November afternoon, United States Vice President Kamala Harris and Meta’s Nick Clegg rush into a large marquee set up in the grounds of a 19th century British mansion.e century, in north London, sit around a round table and, among other things, set out to save the world.
Under the gloomy skies of Bletchley Park, which had been the headquarters of the Allied codebreaking services during the Second World War, Nick Clegg and Kamala Harris join an assembly of international leaders, academics and big business bosses. tech to evoke what some pessimists consider to be a new sword of Damocles over the planet: an uncontrolled surge in artificial intelligence (AI).
After two days of discussions, political leaders and tech luminaries agreed on a joint declaration of good intentions, but did not have a joint proposal on the measures to take. Instead, they present competing plans for how to manage a technology that will be central to the coming decade – and which is expected to upend everything from business to health systems to democracy itself. even.
For those in the front row, the haggling under the English drizzle evokes the division of the world by the great European powers of the 19th century.e century. “We would have thought we were in an alternative reality”, summarizes Amba Kak, director of the AI Now Institute, a non-profit organization invited to take part in the debates. At the end of the meeting, 29 countries, including China, member states of the European Union and the United States, signed a voluntary agreement aimed at reducing risks that have become high on the agenda since the advent from ChatGPT, [le robot conversationnel de la société] OpenAI.
New international standards
For the purposes of this article, Politico spoke with around thirty politicians, decision-makers, tech bosses and others – many of whom wished to remain anonymous on a sensitive subject. The question before them is whether the European Union, the United States or the United Kingdom – or anyone else – will be able to develop a road map that can achieve consensus among Western democracies.
A month before the conference kicked off in rainy England, decision-makers had gone out of their way to move the issue forward on the other side of the world. It was in October. Vera Jourova came out exhausted from a 16-hour flight between Brussels and Japan.
The Czech was only a few weeks into her new role as EU ambassador for technology issues, and her first mission abroad was going to be tough. It consisted of promoting European rules on AI during a G7 summit. The event aimed to develop new international standards around the most advanced form of this technology, “generative AI”, at the origin of ChatGPT and its competitors.
Brussels’ strategy was part of the European law on artificial intelligence, the world’s first attempt at binding legislation in this area. Unlike the position favored by the United States, the EU’s vision includes bans on the most invasive forms of the technology in question and strict rules forcing companies like Google and Microsoft to be more transparent about the design of their products using AI.
“Generative AI has taken over our lives so fast, we need something fast,” declared Vera Jourova to Politico as she exited the high-speed train that had taken her in two hours from Tokyo to the summit of Kyoto.
Brussels, digital policeman
During this three-day meeting in Japan, the American Nathaniel Fick arrived with a competing proposal. A senior digital representative in Joe Biden’s team and himself a former tech boss, Fick did not advocate any bans or strict requirements. Instead, he pleaded
