"I'm sorry Dave. I'm afraid I can't do that" and other news

Some may remember the line above from the movie 2001: A Space Odyssey, a movie that premiered 58 years ago. It was prescient it seems as it was meant to suggest that computing would evolve to a point where the computers would achieve a level of autonomy that would supersede human control and that they would view humanity as an impediment to their mission.

Most will have heard recently that some of the top AI researchers have raised the alarm about artificial intelligence and have caused them, including Canadian researchers Yoshua Bengio and Geoffrey Hinton, to revise their estimates of when the broad cognitive competence of AI systems could be achieved. Previously, estimates of decades or a century have now been revised to suggest it could be a matter of months or a few years before we would have difficulty discerning whether we are interacting with a human or a machine.

Researchers like Jacob Coxon, who quit the AI firm Anthropic, suggested that, "if we don't slow down at the current rate of progress, there is a strong chance we could all die in the immediate future". That sort of prediction tends to get attention. While it seems obvious that advances in the capabilities of AI have outstripped policy, it also raises questions about how to deal with it. This is not a theoretical concern, as there is a sustained race between world powers, most notably the United States and China, to exploit and dominate this technology. The amount of capital being spent on AI is ensuring exponential increases in data centre and compute capacity but it also will see AI be a topic for policy makers and regulators for the foreseeable future, if that still has meaning.

I am not in any way qualified to predict where this will end, but I do believe that reflexive efforts by those for whom broad-based and pervasive regulation is the response to everything should be cautious. Ex ante regulation as the presumptive path in response to new technologies will hobble their development. That, in turn, will leave countries doing so at a competitive disadvantage with their peers. While the spectre of agentic AI gone rogue raised by the founders may be worthy of scrutiny, so too should the motivations of the people making those claims.

Watching the founders of the biggest American AI platforms making their pilgrimage to Washington has made me consider whether their calls for regulation are entirely altruistic. While I am sure they do not desire the extinction of civilization, I equally believe they have an interest in being intimately involved in any regulation that might develop. The regulation they likely would accept, and no doubt are advocating for, is some form of self-regulation and in some areas that might be appropriate. However, as agentic AI evolves and the agents begin to have access to financial and other sensitive personal information, self-regulation may not be the preferred model. For example, providing them with safe harbours or shields from litigation for criminal or tortious acts committed by their AI agents would remove any incentive on the AI platforms to effectively control them. My point in saying this, is that policy makers should not necessarily take the AI founders as acting without self interest, particularly as it concerns what form and scope of regulation ultimately develops.

Investing in a sovereign broadband backbone network connecting Canadians coast to coast with direct links to Europe and Asia is a worthy idea

I was very happy to hear the Prime Minister suggest, in his remarks at the Investment Summit he convened in Toronto two weeks ago and then again in Strasbourg while addressing the European Parliament, that the government intends to invest in a sovereign broadband backbone network. I also applaud the Prime Minister's stated intention in a recent interview he gave to the New York Times, to diversify away from Elon Musk's Starlink satellite service for critical government connectivity and to develop a more formal national satellite strategy.

I wrote about both of these potential developments in my post some eight months ago titled, "Telecom Infrastructure is Security Infrastructure and what might be done after the Speech". Some observers have commented on various interpretations of what they believe the Prime Minister has in mind. Most, in my opinion, have missed the mark by focusing on no doubt worthy, but less ambitious, pursuits like re-routing traffic to Canadian internet exchange points from internet exchanges in the United States. Others have castigated the current incumbents for not investing or underinvesting in economically challenging projects in Canada's North, and still others seem to pair any project to diversify supply, or build a Canadian coast to coast and beyond fibre optic backbone network, as an opportunity to achieve social policy purposes they favour. In my opinion, while those may be laudable objectives, they should follow from and not lead the ambition for such a national project.

I believe, as I wrote in January, that there is an opportunity with the Prime Minister's pledge to increase our NATO commitment to 5% of GDP pledge by 2035 to achieve a number of his sovereign internet ambitions for Canada. 1.5% of the 5% NATO commitment ($43 billion) can be spent on defence-related infrastructure. The government has already made announcements involving the port of Churchill and the Contrecour container facility at the port of Montreal. Directing some the annual $43 billion toward a national fibre optic backbone build would be a worthy and important project. The existing fibre segments are in many cases owned by foreign interests, route traffic through the United States or are in need of investment or replacement. A new Canadian owned and controlled coast to coast fibre network could achieve the Prime Minister's ambition and more. A re-imagined national fibre project could include investments in a sovereign intercontinental fibre project through the Northwest passage for instance, and onward to Europe and Asia, utilizing the shortest physical route.

We find ourselves at a unique point in time where investments in these types of projects are not only necessary but critical to Canada's sovereignty and future prosperity. Hopefully we rise to the opportunity and occasion and proceed with a gimlet-eyed determination to make the most of the moment.