What happens when the AI spark meets the grievance politics kindling?

What happens if new jobs don't replace the ones AI is now vaporising?

What you have happening right now is you have essentially every large company is overstaffed. We could debate how much; it’s at least overstaffed by 25 per cent. I think most large companies are overstaffed by 50 per cent. I think a lot of them are overstaffed by 75 per cent.
Marc Andreessen, 20VC, 30/3/26

We are clear-eyed about the risks—of jobs and entire industries being disrupted; bad actors misusing the technology; misaligned systems evading human control; governments or institutions deploying AI in ways that undermine democratic values; and power and wealth becoming more concentrated instead of more widely shared. Indeed, we highlight these risks here to raise awareness of the need for policy solutions to address them. Unless policy keeps pace with technological change, the institutions and safety nets needed to navigate this transition could fall behind.
OpenAI, Industrial Policy for the Intelligence Age: Ideas to Keep People First, April, 2026

For 30 years, the established parties have let too many people down on affordable and available housing; appropriate, inexpensive and satisfactory education; secure employment and the ability to bargain for better incomes; quality of life and access to services, especially low-cost or free health services; and, generally, the maintenance of a sense of hope and the ability to feel comfortable within society…

In shifting to the [populist] One Nation cause, each voter has to negotiate a series of attitudinal lines relating to Hanson’s stated positions on First Nations people, Asians, those of the Muslim faith, Donald Trump, and the general presence of immigrants in Australian society and whether they should be shipped back to where they came from.

But for many those considerations appear to be subordinate to Hanson’s inchoate message, which is that the modern economy leaves too many behind.
Shaun Carney, SMH, 9/4/26

After close to half a century of neoliberalism, a significant proportion of the population in Anglosphere nations feels aggrieved.

Quantifying exactly what proportion of the population is ready to man the barricades is tricky. But at least a third and possibly well over half the population in nations such as Australia, Canada, New Zealand, the UK and the US believe “the modern economy leaves too many behind”.

(A 2025 IPSOS report found “56 per cent across 31 countries feel society is broken in their country” and 64 per cent agreed with the statement that “traditional parties and politicians don’t care about people like me”.)

You can, as free-market enthusiasts often do, point out that neoliberalism hasn’t been all bad. You can, as progressives often do, lament the grievance politics resentments of the unenlightened lower orders.

At this juncture, all that’s largely beside the point. As was made clear in 2016, a critical mass of Anglosphere voters believe they’ve been shafted.

That’s the kindling. Now let’s move on to the rapidly approaching spark.

AI-driven unemployment
To be fair, I should point out that the Andreessen quote above is taken out of context. It’s part of a longer answer where he claims the redundancies we’re seeing at companies such as Oracle are a result of over-hiring during or after the Covid pandemic.

Andreessen is certain there’s nothing much to fear from AI-driven automation and new jobs will appear to replace the old ones. (He doesn’t appear to dispute the claim AI will create significant labour market displacement, but is confident it will create so many new, yet-to-be-imagined roles that mass unemployment won’t be an issue.)

I want to stay on topic, so I’ll just point out that I haven’t yet seen any media reports of businesses hiring thousands of people to fill all the new roles AI has created. Nor have I heard any CEOs declaring they’ll need to dramatically expand their workforces so their company can thrive during the AI era. And now I think about it, apart from ‘AI ethicist’, I can’t identify any new job categories AI has created since ChatGPT dropped in late 2022.

So, as well as being enraged about cost-of-living pressures, unaffordable housing, rapid demographic change, out-of-touch elites and the absence of “a sense of hope and the ability to feel comfortable within society”, many Anglosphereans now fear they may lose their job and be unable to secure alternative employment.

Ideas to keep people first
For the sake of argument, let’s assume AI does put a lot of people out of work.

The danger is not mass unemployment breaking out overnight. It’s likely to be a slower process – ever fewer entry-level roles, ever more automation of routine cognitive work and ever more downward pressure on wages for those who manage to remain employed.

It’s difficult to predict what happens when unemployment rates start climbing, but it’s unlikely to end happily for AI company founders or other members of the elite.

So, it’s unsurprising that AI company C-suiters, among other high-powered, well-remunerated intellects, have been giving serious thought to how AI can transform economies without setting off a revolution led by populists of the Left or Right. Or, who knows, perhaps a unity ticket of Left and Right populists.

Which brings us to OpenAI’s recent thinkpiece, which “offers initial ideas for an industrial policy agenda to keep people first during the transition to superintelligence”.

The document starts with the usual throat clearing about how AI promises enormous upsides (“superintelligence will speed up scientific and medical breakthroughs, significantly increase productivity, lower costs for families by making essential goods cheaper, and open the way for entirely new forms of work, creativity, and entrepreneurship”) while also posing some challenges (a turbocharging of the concentration of wealth and power that’s occurred throughout the Anglosphere since the Keynesian settlement was abandoned in the 1980s).

The way to square this circle, according to OpenAI, is to navigate the AI transformation process “through a democratic process that gives people real power to shape the AI future they want, and prepare for a range of possible outcomes while building the capacity to adapt”.

As the OpenAI authors note, “This shift will reshape how organizations run, how knowledge is created, and how people find meaning and opportunity” while highlighting “the limitations of today’s policy toolkit” and raising the risk of “governments or institutions deploying AI in ways that undermine democratic values; and power and wealth becoming more concentrated instead of more widely shared”.

To avoid a dystopian future, OpenAI proposes building an open economy and a resilient society.

What an open economy looks like
OpenAI warns that AI will disrupt jobs unevenly, with some roles disappearing, others changing, and inequality widening unless policy catches up.

Its proposed response is an “open economy” agenda built around worker voice, wider access to AI, and broader sharing of AI-driven wealth.

OpenAI proposes giving workers formal input into workplace AI deployment, helping AI-enabled entrepreneurs launch firms, treating access to AI as a basic economic right, modernising the tax base as labour income becomes less central, and creating a public wealth fund so citizens share in AI gains.

It also calls for stronger safety nets, portable benefits, and significant investment in care work, energy infrastructure and AI-enabled scientific discovery.

What a resilient society looks like
As AI becomes more deeply embedded, it will create new vulnerabilities, including cyber and biological misuse, harms to mental and social wellbeing, systems acting beyond human intent, and strain on democratic institutions.

OpenAI argues resilience can’t rely only on upstream safeguards such as model testing and regulation before deployment. Societies also need systems that work after deployment, in messy real-world conditions.

The agenda OpenAI suggests includes building safety systems for emerging risks, creating an “AI trust stack” for verification and accountability, developing auditing regimes for frontier models, and preparing containment playbooks for dangerous systems that cannot easily be recalled.

It also calls for mission-aligned governance at frontier AI firms, strict guardrails on government use of AI, stronger mechanisms for public input, incident-reporting systems, and international information-sharing to detect risks early and coordinate responses quickly.

A beautiful vision
It’s difficult to fault OpenAI’s recommendations. The question is whether any of them are politically (and geopolitically) feasible. What OpenAI is proposing is, well, revolutionary, especially in societies that have spent decades gulping down the free-market-worshipping, wealth-creator-venerating Kool-Aid.

Then again, elites have chosen to throw the masses a bone in the past, if only out of self-interest. As OpenAI notes,

Following the transition to the Industrial Age, the Progressive Era and the New Deal helped modernize the social contract for a world reshaped by electricity, the combustion engine, and mass production. They did so by building new public institutions, protections, and expectations about what a fair economy should provide, including labor protections, safety standards, social safety nets, and expanded access to education.

History shows that democratic societies can respond to technological upheaval with ambition: reimagining the social contract, mediating between capital and labor, and encouraging broad distribution of the benefits of technological progress while preserving pluralism, constitutional checks and balances, and freedom to innovate.