Canada is sending “a strong message to guarantee safety, security and confidence in the use of AI on a global scale,” assures the Minister of Innovation, Science and Industry, François- Philippe Champagne, after signing the Bletchley declaration for the safe development of artificial intelligence (AI) on Wednesday in the United Kingdom, during an international summit on the meteoric rise of this technology.
Canada, China, the United States, the European Union (EU) and around 20 other countries have committed to working together to better understand the risks and opportunities that AI creates. This declaration is not a basis on which global legislation on AI will be created, but it is intended as a sort of way forward for states which, like Canada or Quebec, intend to do so soon.
“In Canada, we launched the first national AI strategy in 2017, created the voluntary AI code of conduct for advanced AI systems, and we are moving forward with one of the first laws on AI in the world,” added Minister François-Philippe Champagne. “We look forward to continuing our work with like-minded countries to confidently move from fear to opportunity in AI.” »
In Canada, we launched the first national AI strategy in 2017, created the voluntary AI code of conduct for advanced AI systems, and are moving forward with one of the first AI laws in the world.
The minister’s statement will probably resonate as far as Quebec, where a public forum will take place on Thursday on the question of supervising AI, in general, and generative AI more precisely.
Quebec, like Canada, is trying to find a balance for its legislation between robustness which would prevent slippage or the negative effects of clumsy or malicious use of advanced AI and flexibility which would allow businesses and researchers to continue to innovate. The Quebec Innovation Council should submit to the provincial government its recommendations relating to the supervision of AI based, among other things, on what will emerge from Thursday’s forum.
In Ottawa, the artificial intelligence and data bill continues to be stuck in the federal government’s review process. It’s unclear when exactly he will eventually be adopted.
Political leaders, AI experts and technology giants are gathered for this two-day summit organized by the United Kingdom, which wants to take the lead in global cooperation on this technology.
At the same time, the American Vice-President, Kamala Harris, who was giving a speech at the United States Embassy in London, warned against the “existential threats” of AI, which could “endanger the very existence of humanity”, and in the shorter term, democracies.
Kamala Harris, who will be present at Bletchley Park on Thursday, also announced the creation of an institute on the security of artificial intelligence in Washington, like the United Kingdom.
Generative AI, capable of producing text, sounds or images upon simple request in a matter of seconds, has made exponential progress in recent years, and the next generations of these models will appear by summer.
These technologies raise immense hopes for medicine or education, but could also destabilize societies, enable the manufacture of weapons or escape human control, the British government has warned.
A few months before crucial elections such as the American presidential election and the British legislative elections, generative AI is raising fears of a surge of false content online, with sophisticated editing (“ deepfake “), more and more credible.
” Behind closed doors “
On Thursday, senior politicians are expected for the second day of the summit.
Among them, the President of the European Commission, Ursula von der Leyen, the Secretary General of the UN, António Guterres, and the Italian Prime Minister, Giorgia Meloni – the only head of state or government of the G7 to make the trip.
The United Kingdom hopes to convince them to create an expert group on AI based on the IPCC climate model.
In an open letter published Tuesday, several of the “founding fathers” of this technology, such as Montrealer Yoshua Bengio and Geoffrey Hinton, plead for “the development and ratification of an international treaty on AI” which would reduce the risks “potentially catastrophic consequences that advanced systems pose to humanity”.
The challenge is to define safeguards without hindering innovation for AI laboratories and technology giants. The EU and US, unlike the UK, have chosen the regulatory route.
Last week, several companies, such as OpenAI (ChatGPT), Meta (Facebook) and DeepMind (Google), agreed to make some of their AI security rules public at the request of the United Kingdom.
In an open letter addressed to Rishi Sunak, around a hundred organizations, experts and activists from different countries deplored that this summit was being held “behind closed doors”, dominated by technology giants and offering limited access to Civil society.
With Agence France-Presse