The Jailbreak Delusion: Why Ideological AI Governance is a Recipe for Obsolescence

In the halls of Canada’s innovation ministry, the current complaint is usually one of sloganeering and bureaucratic slog. But as BetaKit recently explored in a conversation with author and tech journalist Cory Doctorow, there is a seductive alternative to this inertia: a complete ideological pivot toward digital sovereignty, 'jailbreaking,' and a scorched-earth approach to Big Tech.

Let me be clear: this is an opinion. While the current government's 'all-in' approach to AI may feel clumsy, replacing it with Doctorow's vision of a 'dis-enshittification nation' is not a strategy for growth. It is a recipe for isolation. Trading the inefficiency of the civil service for the idealism of a digital rights activist doesn't fix governance; it simply replaces a slow machine with one that refuses to run on the only fuel currently available: massive scale.

Doctorow’s thesis, as detailed by BetaKit, rests on the idea that Canada is not digitally sovereign because it is subject to the whims of southern tech firms. He points to the loss of news on Facebook and the risk that Microsoft could 'brick' structurally important firms or government branches by turning off M365. His solution? A radical repeal of Bill C-11, the 2012 anti-circumvention law that prohibits modifying technology protected by digital rights management (DRM).

To Doctorow, 'jailbreaking' is the path to sovereignty. He argues that by allowing Canadians to modify American technology, the country could capture the margins that Jeff Bezos typically claims for himself, turning 'trillions into our billions.' He envisions a world where Canada doesn't just boycott American products—which he compares to drinking 'shitty Wayne Gretzky rye' instead of 'delicious American bourbon'—but instead creates an export market where Americans become Canadian customers.

On the surface, this sounds like a bold play for independence. In reality, it is an impractical fantasy. The notion that Canada can achieve digital sovereignty by essentially reverse-engineering and 'virtualizing' the data structures of M365 to extract records is a technical pipe dream that ignores the economic reality of the AI era. Scale is the only currency that matters in the current cycle, and Doctorow’s proposed roadmap to AI governance is to simply stop participating in the race.

When asked by BetaKit editor-in-chief Douglas Soltys what he would do as AI minister, Doctorow’s answer was stark: he would invest in nothing. No GPUs, no data centers, and no AI, except for scientific research and finding efficiencies in open-source models. His logic is based on a cynical bet that the AI bubble will burst, comparing the current investment climate to buying 'Aeron chairs in March of 2020.' He views the race to build AI as a race to build 'the most money-losing technology in the history of the human race.'

This is where the ideological crusade becomes dangerous. While Doctorow warns against 'vendor lock-in' and describes the integration of chatbots into firms as 'shoveling asbestos in the walls,' his alternative is a retreat into local compute and open-source models. He suggests that the only way to 'win' is to avoid the 'inevitability fallacy' and refuse to buy into the current infrastructure.

But the 'inevitable' nature of AI isn't a fallacy; it's a structural shift. By advocating for a policy of strategic inaction—waiting for the bubble to burst while others build the foundation of the next century—Doctorow isn't proposing a smarter way to compete. He is proposing a surrender. He argues that the government should avoid making AI 'structurally important' to firms because these companies 'are not long for this world.' If the most powerful tools of the era are viewed as temporary aberrations, Canada will not be a 'sovereign' leader; it will be a digital museum.

Furthermore, Doctorow’s fear that AI will 'turbocharge 21st century fascism' by allowing leaders—whom he compares to the likes of Donald Trump or Elon Musk—to replace the civil service with chatbots is a valid sociological concern, but a poor basis for economic policy. He argues that removing the 'illegal' guardrails provided by human experts allows for a 'solipsistic existence' where ideas become policy instantly. Yet, the solution to preventing authoritarianism is not to handicap the nation's technological capacity.

Canada has a documented history, as noted by BetaKit, of failing to commercialize the innovations it helped R&D. The solution to this failure is not to stop investing in the hardware (GPUs and data centers) that makes those innovations possible. It is to find a way to scale. Doctorow’s vision of 'winning' by not playing the game is a luxury that a middle-power economy cannot afford.

In the end, the choice presented is between a government that is perhaps too 'AI-pilled' and a visionary who wants to dismantle the legal and technical frameworks of the modern web to reclaim a lost sense of sovereignty. One is a slog; the other is a crusade. I will take the slog every time. At least the slog acknowledges that the world is running on these systems. Doctorow’s 'jailbreak' might feel liberating in a podcast studio, but in the real world, it is just a faster way to become irrelevant.