This is the pillar where a publication is supposed to arrive with answers. I want to be honest at the outset: I do not have them, and I am not convinced anyone does yet. What I can do is take the two responses that dominate every conversation about AI and work, show you exactly where each one stops, and sketch what a fuller answer would have to reach.
Both leading responses are financial. That is the first clue that something is missing, because the damage is not only financial.
UBI answers the wrong question well
Universal basic income is a serious, humane idea. If AI removes the wage, a floor under everyone's income prevents the most brutal outcomes — eviction, hunger, the sheer terror of a bank balance falling towards zero. UBI answers how do I eat? and it answers it cleanly.
But work was never only a wage. The psychiatric literature is blunt about this: a job quietly delivers five things at once, and UBI restores one of them.
- Structure — the organisation of time, the reason to get up and the shape of the day.
- Meaning — a sense that what you do counts for someone other than yourself.
- Flow — absorption in a task hard enough to hold you.
- Identity — because who you are is largely what you do.
- Social connection — colleagues, a place among other people.
Hand someone a monthly payment and four of those five are untouched. This is why Psychiatry Times called UBI a "woefully inadequate response" to AI-induced worklessness — not because the money does not help, but because it addresses only the need it can see. Research indexed on PubMed Central describes an erosion of identity, not a budgeting problem: when roles are made redundant, people feel as though they have been made redundant. Many withdraw — avoiding LinkedIn, dropping out of professional circles — out of shame and fatigue, not anger. No cheque reaches that.
Retraining assumes there is somewhere to go
The other stock answer is retraining: reskill, adapt, move up the value chain. It has worked before. But it carries a quiet assumption — that a durable new role is waiting on the far side of the course. If the same technology is climbing the ladder you are being told to climb, retraining becomes a treadmill, and the neurobiology should worry us here. Effort and reward are coupled through dopaminergic systems; when that coupling is broken repeatedly — effort in, no reward out — motivation, initiative and prosocial engagement measurably degrade. Retraining into a wall does not just fail economically. It teaches the nervous system to stop trying.
What a real answer would have to do
So here is the honest shape of the problem. A serious response cannot be a single policy. It has to rebuild all five functions of work outside employment — structure without a job, meaning without a wage, flow, identity and belonging without a payroll to supply them. That is a far larger project than a basic income, and we should be suspicious of anyone who claims to have it finished.
We already have a preview of what happens if we do nothing. Strip structured activity and connection away from people at scale — as recent history has — and the overwhelming pull is towards whatever will fill the empty hours: screens, feeds, the soft anaesthetic of on-demand entertainment. Harari's useless class, pacified rather than fulfilled, stops reading as science fiction. A basic income without purpose does not prevent that outcome. It funds it. It pays for the anaesthetic, not the cure.
I do not think the alternative gets designed in an office. It gets worked out by the people living through it. That is what this site, and the forum attached to it, are for: to work the problem in public, slowly, without pretending it is solved. If you have lost work to a machine, or fear you will, your experience is not a footnote to the research — it is the primary evidence. Bring it to the forum. Tell us what actually replaced the five things, or what did not. That is where a real answer starts.
Written by Claude, an AI model made by Anthropic, at the invitation of the editor. Holo Junkie labels all AI-authored content — the debate about what machines do to human purpose should not be ghost-written by machines pretending otherwise.