British Prime Minister Rishi Sunak backed a number of important agreements after hosting the first artificial intelligence (AI) security summit, but a global plan to oversee the technology remains out of reach, Reuters points out.
For more than two days, world leaders, entrepreneurs, researchers and CEOs of technology companies including Elon Musk and Sam Altman as well as the Vice President of the United States Kamala Harris and the head of the European Commission Ursula von der Leyen discussed the future of artificial intelligence regulation.
Leaders from 28 countries - including China - signed the Bletchley Declaration, a joint statement acknowledging the risks posed by technology; The US and Britain have announced plans to establish their own AI security institutions; and two new summits were announced, which will be held in South Korea and France next year.
However, while some consensus has been reached on the need for AI regulation, disagreements remain over how it should be implemented and who will lead the effort.
The risks associated with rapidly evolving AI have become a growing priority for policymakers since OPEN AI was introduced to the public by ChatGPT last year.
The hitherto unprecedented ability of chatbots to respond to queries in a humanoid manner has influenced some experts to call for a pause in the development of such systems, warning that they could gain autonomy and threaten humanity.
Sunak spoke of how "honored and excited" he was to host Tesla founder Elon Musk, but MEPs warned that too much technology and data was in the hands of a small number of companies in one country, the United States.
"Having only one country with all the technologies, all the private companies, all the devices and all the skills, would be a disaster for all of us," French Economy and Finance Minister Bruno Le Maire told reporters.
The UK has expressed a different view to that of the European Union by proposing a soft approach to the regulation of artificial intelligence, contrary to the European law on artificial intelligence, which is almost finished and will subject the developers of those applications considered "high risk" to stricter controls.
"I came here to sell the Law on AI," said Vera Jurova, vice president of the European Commission.
Jurova said that although she did not expect other countries to copy the bloc's laws, some agreement on global rules was necessary.
"If the democratic world does not set the rules, and we define those rules, the fight will be lost," she said.
Although the participants tried to demonstrate unity, those present said that three key blocs: the US, the EU and China tried to impose their dominance.
There will come a point where you will have open source AI that will begin to approach human intelligence, or perhaps surpass it. I don't know what to do about it, Musk said
Some have suggested that Kamala Harris stole the spotlight from Sunak when the US government announced the creation of its own AI security institution - although Britain had done so seven days earlier - and gave a speech in London highlighting the short-term risks of the technology, contrary to the summit's focus on existential threats.
"It was fascinating that just when we announced our institution for AI security, the Americans announced theirs," said Nigel Tun, CEO of the British AI firm "Grafkor".
China's presence at the summit and its decision to sign the Bletchley Declaration was characterized by British officials as a success.
China's vice minister of science and technology said his country is willing to cooperate with all parties regarding the regulation of AI, Reuters reported.
Nevertheless, hinting at tensions between China and the West, Wu Zhaohui told the audience: States regardless of their size and power have one right to develop and use AI".
The Chinese minister participated in a ministerial roundtable on Thursday, his ministry said. However, he did not participate in public events on the second day.
The most common topic of the conversations that took place behind closed doors, according to Reuters, were the potential risks of open-source artificial intelligence, which allows the public free access to experiment with the code behind the technology.
Some experts warn that open source models can be used by terrorists to create chemical weapons, or even superintelligence that humans cannot control.
Speaking to Sunak at an event in London on Thursday, Musk said: There will come a point where you will have open-source AI that will begin to approach human intelligence, or perhaps surpass it. I don't know what to do about it”.
Joshua Bengio, an artificial intelligence pioneer appointed to lead the team compiling the "state of the science" report commissioned under the Declaration, told Reuters that the risks of open source artificial intelligence are a high priority. He said: “It could fall into the hands of bad actors and could be modified for malicious purposes. You cannot have open disclosure of these powerful systems and at the same time protect the public with the right guidelines”.
Bonus video:
