November 3, 2023
SINGAPORE – Countries need to work together to better understand ethical artificial intelligence (AI), to develop and deploy frontier AI, and shape its safeguards, Prime Minister Lee Hsien Loong said, after attending the Artificial Intelligence Safety Summit (AISS) on Thursday.
“The field of artificial intelligence is developing rapidly, transforming lives while raising deep ethical questions. We grappled with these emerging issues in our Leaders’ session,” PM Lee said on Facebook, after attending the summit virtually.
The two-day summit in Britain took place on Wednesday and Thursday at Bletchley Park – a former code-breaking spy base.
It brought together governments, leading AI companies, civil society groups and research experts to discuss the risks of AI, especially at the frontier of development, and how such risks can be mitigated through internationally coordinated action.
PM Lee was invited by British Prime Minister Rishi Sunak, and spoke about Singapore’s approach to AI development and its usage.
On better understanding ethical AI, PM Lee said: “Be it decisions made by self-driving cars or doctors relying on AI-generated diagnoses, the AI systems must be imbued with human context and human values.
“We welcome the UK’s new AI Safety Institute and its cooperation with Singapore on safety testing,” he added, also saying that Singapore has introduced measures, like the AI Verify testing toolkit, to mitigate these risks.
On frontier AI, PM Lee said that countries need to work together for mutual security, even if competition is inevitable. Frontier AI refers to advanced AI systems that are still in their early stages of development, but have the potential to revolutionise many industries and aspects of people’s lives.
“No one benefits when AI systems go rogue. Countries will ultimately need to establish some global understanding to make AI systems safer, and avoid AI creating strategic risks and instability,” PM Lee said on Facebook.
He added that all stakeholders should participate in shaping the rules and safeguards that govern AI. This is regardless of the size of the country, since smaller nations will also be affected by AI.
“While the main players are American, Chinese and European, this conversation on AI safety cannot just be amongst the few,” he said. “Small countries like Singapore, too, participate in AI research and deployment, and we too will certainly be affected by AI, including both its benefits (and) its risks and downsides.
“Singapore is honoured to work with international partners so that we can all reap the benefits of AI, and make AI a force for good contributing to our common prosperity.”
Singapore was among 28 countries invited to contribute views on the matter, with the Republic represented by Communications and Information Minister Josephine Teo.
Other attendees included United States Vice-President Kamala Harris and United Nations Secretary-General Antonio Guterres.
On Wednesday morning (British time), Mrs Teo chaired a round-table discussion on whether highly advanced AI systems could lead to the loss of human control and oversight in the future, along with the risks this posed, and how to prevent such a scenario.
In her closing remarks, she said: “Current AI systems do not yet pose a real risk of ‘loss of control’. They require human prompting, and generally fail when asked to plan over time towards a goal.
“However, future models are likely to improve on these dimensions. Even when AI systems appear to display high cognitive abilities, we cannot be sure that they will behave like, or take the same decisions as, humans.”
She said in a LinkedIn post that the participating countries agreed on three goals following the discussion.
The first is to develop global expertise in the safety research and development of AI. The second aim is to deepen collaborations in AI testing and auditing, and the third goal is to continue exchanges among various parties, including governments, industries, academia and civil society.
“Ultimately, we want to co-create trustworthy AI that can effectively benefit (the) economy and society,” she said in her closing speech.
In addition, the delegates from the different participating nations agreed on Wednesday to collaborate to contain the potential risks that cutting-edge advancements in AI posed, a move that Mr Sunak described as a “landmark achievement”.