Canada announces a $2.3 billion AI strategy following a papal suggestion on safety measures.

      TL;DR: Canada has introduced “AI for All,” a $2.3 billion national AI strategy centered on six pillars, including sovereign infrastructure, job creation, and AI literacy. Prime Minister Carney announced this alongside a conversation with Pope Leo XIV regarding responsible AI, but critics point out the strategy lacks specific safety timelines.

      Just days after discussing the ethical implications of artificial intelligence with Pope Leo XIV, Canadian Prime Minister Mark Carney revealed a national framework in Toronto that aligns with the pontiff's requests. Named “AI for All,” the strategy allocates over $2.3 billion in funding over the next five years, marking Canada’s most ambitious initiative to become a prominent player in the global AI arena.

      However, the strategy notably lacks essential safety measures, offering little in the way of concrete protocols, timelines for new regulations, or enforcement mechanisms. The Pope cautioned that AI should be “disarmed” by governments.

      The six pillars of the plan, initially presented in the spring economic update, include: protecting Canadians and democracy, empowering citizens, fostering shared prosperity, establishing a sovereign AI foundation, enhancing Canadian champions, and forming trusted partnerships and global alliances.

      Minister of Artificial Intelligence Evan Solomon stated that the strategy reflects the desires of Canadians. “Canadians want safe, reliable, and sovereign AI,” he emphasized. “They seek the best tools to foster a prosperous future aligned with our values.”

      The strategy prominently features job creation, aiming to generate up to 90,000 AI-related positions for young Canadians by 2031, along with an additional 250,000 roles through broader AI adoption in the economy. The goal is to elevate AI usage in businesses from about 12% to 60% by 2034.

      While safeguarding sovereignty is a key focus, safety measures take a backseat. Canada’s heavy reliance on foreign cloud services is characterized as a vulnerability, prompting a “build-partner-buy” strategy: developing essential capabilities within Canada when feasible, collaborating with trusted allies, and acquiring from the market as needed.

      Concrete initiatives include establishing a “world-leading” supercomputer and creating sovereign data centers capable of delivering 100 megawatts to Canadian clients, with $1 billion earmarked for public supercomputing infrastructure.

      This sovereignty initiative builds on an alliance already initiated by Carney’s administration. In February, Solomon and his German counterpart signed a joint AI declaration during the Munich Security Conference, establishing a Sovereign Technology Alliance aimed at decreasing reliance on dominant tech providers. The strategy outlines plans to expand this alliance, drawing parallels with Europe’s focus on technological sovereignty.

      Concerning the papal influence, Carney and Pope Leo discussed AI's moral implications on May 29, shortly after the release of the Pope's encyclical, Magnifica Humanitas. This document, published on the 135th anniversary of Rerum Novarum, identifies AI as this era's industrial revolution and raises concerns about its potential to exacerbate inequality, diminish human agency, and centralize power if not appropriately regulated.

      Both the Vatican and the Prime Minister's Office reported that the conversation included the necessity for AI to benefit humanity, starting with individual protection. Carney expressed Canada's intent to take a leading role internationally in fostering responsible AI development.

      The strategy mentions forthcoming consumer privacy legislation that will establish privacy rights and protect children's online information, alongside a commitment to update safety laws, though it lacks definitive timelines for implementation.

      Earlier this year, Minister Marc Miller raised the possibility of prohibiting AI chatbots for those under 16, but this measure is not included in the strategy. Officials indicate it is still under review and may be incorporated into upcoming online harms legislation.

      This omission is particularly significant in light of the global landscape. The EU is advancing its own sovereign AI infrastructure while enforcing the AI Act, the most comprehensive regulatory framework worldwide. Canada’s past attempt at AI legislation, the Artificial Intelligence and Data Act within Bill C-27, is considered inadequate and has not been revisited.

      On the workforce front, the strategy introduces a national AI literacy initiative that offers free introductory training for all Canadians. Ottawa aims to reach one million post-secondary students and train over 3,000 educators with AI-focused learning resources. Additionally, $30 million will be allocated to CanCode, a federal program funding non-profits that provide digital skills training to youth. The strategy also intends to expand the Global Talent Stream permit program to facilitate entry for highly skilled AI professionals, although specific visa targets remain unspecified.

      The ongoing debate about the jobs AI will generate versus those it will eliminate continues. The projected 250,000 new positions lack an established methodology, and independent economists have not yet assessed this claim.

      As for the future, cross-party pressure is mounting, with some parliamentarians advocating for a complete halt to the development of superintelligent AI, emphasizing that safety measures should precede industrial strategies. This perspective coincides with the Vatican's stance, where Pope Leo advocated for AI to be “disarmed” and rejected the notion of