Dhe name of the southern English country estate Bletchley Park has a mysterious ring to it. During World War II, several thousand military intelligence officers worked in the complex. Mathematicians and computer scientists deciphered the Enigma code of the German radio messages. Alan Turing was a key figure here. The old brick mansion, which was then the “top secret” secret service headquarters, still stands today. A little further away, behind barbed wire, you can see flat gray barracks, some with broken windows, where code breakers used to sit.
Like Enigma back then, artificial intelligence (AI) is now considered a difficult puzzle to solve. And many worry that super-intelligent computer programs could not only bring great benefit, but also cause great harm.
Now the British government has invited people to a major summit meeting in Bletchley Park, north of London. Government representatives from 28 nations are coming, US Vice President Kamala Harris, EU Commission President Ursula von der Leyen and other government representatives from Europe, but also China’s Technology Vice Minister Wu Zhaohui, India’s IT Minister Rajeev Chandrasekhar and government representatives from Saudi Arabia and Nigeria.
Race for the strongest AI programs
In a video message to summit participants, King Charles compared the new technology to the discovery of electricity or the splitting of the atom. AI could be a “force for good” – but it could also endanger civilizations and democracies. Experts see possible dangers from disinformation, election manipulation or cyber attacks.
The fact that the British are bringing together a very wide variety of nations with high-ranking political representatives and top corporate executives for the AI summit for the first time is considered a success in London. Science Minister Michelle Donelan spoke of a “truly historic moment”. The most prominent guest from the tech industry was multi-billionaire and Tesla owner Elon Musk, who is involved with the AI start-up xAI. The heads of ChatGPT parent Open AI and Google’s Deep Mind are also there. The Chinese participation in the confidential Bletchley Park debates on AI developments caused some nervousness in diplomatic circles.
The race for the strongest AI programs not only promises great economic potential and solutions to human problems, for example in medicine or education. It could also trigger geopolitical tensions. It would be all the more important if international rules for AI could be agreed upon. British Prime Minister Rishi Sunak has previously joined those calling for an international body on AI risks similar to the UN Climate Change Council (IPCC). Great Britain is setting up its own institute for AI safety.
“Existential risk for humanity”
But whether the conference in Bletchley Park was really the big breakthrough remained questionable. The government in London downplayed expectations in advance. A two-page final communiqué, the “Bletchley Declaration,” was published on Wednesday. Prime Minister Sunak spoke of a “milestone”. The 28 participating nations confirm that they understand the risks and want to manage them “collectively”.
AI offers enormous global opportunities for prosperity and peace. But AI could also cause “serious, even catastrophic damage,” intentionally or unintentionally, the statement said, referring to cyberattacks or disinformation. The signatories advocate international cooperation and announce further AI security summits. However, the paper does not contain any more concrete approaches for international, even global, regulation.
Federal Digitization Minister Volker Wissing then said that a uniform competitive environment and a “level playing field” were needed. AI has a huge influence on the competitiveness of the German economy. You have to play at the front. Wissing called for a coordinated discussion about risks. “There should be no competition for the strictest regulation,” he said. Europe could coordinate closely with American regulatory approaches. It must build on values such as transparency and democracy.
Particularly drastic demands had previously been made by a group of Chinese experts led by the prominent computer scientist Andrew Yao, who had published a dark appeal before the Bletchley summit. AI has the potential to become a “risk to the existence of humanity” in the coming decades. They therefore called for the establishment of an international regulatory authority. The most modern AI systems would have to be registered and tested there. And they should include a “shutdown” mechanism to stop them in an emergency. The AI industry should spend 30 percent of its research budget on AI security, the group demands.
On the other hand, lobbyists from tech companies took a position before the Bletchley summit and warned against too much and hasty regulation, which threatened to stifle research and innovation. Nick Clegg, chief lobbyist for Meta, the Facebook parent company, warned before the summit against over-regulation of rapidly developing technologies. He sees a competition between those who paint existential risks from AI on the wall. Gloomy prophecies would be published breathlessly. This has also been the case with video games, radio and the Internet. You shouldn’t put the cart before the horse with regulation, said former British deputy prime minister and current meta-lobbyist Clegg. First you should look at where the AI programs went, then you can contain them.
#Risk #Summit #Force #Good #Catastrophic #Damage