at the paris artificial intelligence action summit in february, then prime minister justin trudeau and other leaders watched united states vice president jd vance attack artificial intelligence regulation, a moment that helped trigger a wider global shift in attitudes toward artificial intelligence governance. that shift reached canada when mark carney replaced trudeau as prime minister and signalled a move away from the trudeau government’s ambition to be first with artificial intelligence regulation legislation toward a more growth focused approach. under the liberals, then industry minister françois philippe champagne had advanced bill c 27 to update private sector privacy law and impose obligations on “high impact” artificial intelligence systems, but carney’s new artificial intelligence minister evan solomon arrived promising not to “over index” on artificial intelligence rules, arguing canada would not regulate alone if the united states and china stayed away from governance.
over less than a year, ottawa’s focus shifted from fencing in artificial intelligence to chasing its economic opportunities and driving adoption in the federal public service, even as solomon signed new artificial intelligence agreements with germany, the united kingdom and the european union at the g7 industry, digital and technology ministers’ meeting in montreal. solomon insisted canada’s stance had not changed and described a “sweet spot” between what he called european over regulation that constrained innovation and the approaches in the united states and china, while european commission executive vice president henna virkkunen stressed that both sides shared a human centric, democracy based vision for technology. both officials portrayed united states resistance to artificial intelligence regulation as nuanced and highlighted state level efforts, a view complicated days later when united states president donald trump signed an executive order blocking united states states from regulating artificial intelligence. virkkunen said canada would “decide in the coming months … what kind of rules canada would like to set,” as solomon prepared a new privacy bill that will revive the privacy elements of bill c 27 but leave out its broad artificial intelligence regulatory regime, focusing instead on protecting children and addressing deepfakes, potentially including age limits for chatbots and rights to delete synthetic content.
experts say carney’s government is prioritizing artificial intelligence as a commercial and nation building platform while keeping public investment relatively modest and delaying detailed governance choices. university of british columbia professor heidi tworek said the trump administration’s stance makes passing regulation harder and that the carney government is more concerned with innovation, which she sees reflected in an artificial intelligence task force criticized for being heavily weighted toward industry voices and in carney’s push for “sovereign artificial intelligence” and major projects. centre for international governance innovation president paul samson said carney is looking to artificial intelligence to boost canada’s weak productivity, noting “he needs economic growth and productivity growth, and artificial intelligence is an area that can deliver some of that, which is hard to come by these days,” while western university chief artificial intelligence officer mark daley called it “absolutely essential” that canada appointed an artificial intelligence minister and argued ottawa understands that artificial intelligence and compute are “the railroad of the twenty first century.” yet carney has not matched the rhetoric with large new spending, as the 2024 budget under trudeau allocated $2.4 billion to artificial intelligence development, largely for compute and sovereign artificial intelligence infrastructure, while carney’s first budget in fall 2025 included $925.6 million for sovereign artificial intelligence infrastructure to increase compute availability, of which $800 million came from previously allocated funds.
carney has also floated building a “sovereign cloud” as part of his major projects agenda, but none of the government’s announced projects so far are focused on actually constructing one, leaving observers waiting to see whether funding will change after solomon unveils an updated national artificial intelligence strategy in the new year. samson said many countries would welcome canada taking a stronger role in international artificial intelligence governance, such as efforts to bar artificial intelligence from nuclear weapons systems, but argued ottawa is “a little bit shy” because of united states politics and the power of big technology firms. he said representatives of non g7 countries and major international organizations have privately encouraged canada to step up as a catalyst that is “a little more plugged in, like a g7 country,” since only a small number of states can credibly convene the kind of broad, sensitive discussions needed on global artificial intelligence rules. for now, canada’s artificial intelligence posture remains defined by a clear political embrace of economic opportunity and sovereign capacity, paired with a cautious, still in progress approach to concrete regulation and international leadership.
