Anthropic recently published a policy paper about AI, jobs, and what governments should do if AI causes major labor-market disruption. The original paper is here: Anthropic’s economic policy proposal.
The paper is about the question: if AI creates huge wealth while also replacing a lot of human labor, who gets the money, who pays the costs, and who gets blamed?
Anthropic is warning that AI could seriously disrupt jobs. It says governments should prepare now with better unemployment systems, wage support, retraining, public benefits, and possibly new taxes or wealth-sharing mechanisms later. That sounds responsible. It also protects Anthropic’s business interests.
Anthropic’s main argument is that AI may create massive economic growth, while also reducing demand for human labor, so AI might make the economy richer while making many workers less necessary.
The company says society should prepare for that possibility. It also says society should try to preserve work and dignity, rather than only giving people financial support after the damage is done.
Anthropic proposes a staged response.
Stage 1
In the first stage, unemployment still looks normal, but the job market starts changing faster. People may lose jobs, switch careers, or find that their skills become outdated more quickly.
Anthropic suggests wage insurance, training, job-matching tools, licensing reform, and “capital accounts” that could give people some ownership in the AI-driven economy.
These ideas sound reasonable while being slow, complex, and politically difficult. Meanwhile, Anthropic and other AI companies can keep growing quickly.
Stage 2
In the second stage, unemployment rises to recession-like levels. At that point, Anthropic says governments should expand unemployment insurance, provide more direct support, help people move between sectors, and make sure basic needs are covered.
So if AI companies create tools that replace workers, the immediate solution in the paper is mostly public support: unemployment insurance, government benefits, training programs, and emergency relief.
That means taxpayers and governments carry the burden, while AI Companies get the profits.
Stage 3
The third stage is the most serious one. This is the world where AI becomes a broad substitute for human labor. The economy might be producing more than ever, while many people struggle to earn a stable income through work.
At that point, Anthropic says society may need bigger changes: new taxes, public investment funds, AI-use levies, digital dividends, universal basic income, worker equity-sharing, or other ways to redistribute the gains from AI.
Anthropic is basically admitting that if AI changes the economy at scale, existing welfare systems will be too small for the problem.
The private upside is happening now. The redistribution is left for later. And “later” probably means politics, lobbying, negotiation, delays, and loopholes.
Why Publish It?
The positive reading is that Anthropic sees a real risk and wants governments to prepare.
The marketing reading is that Anthropic wants to be seen as the responsible AI company. The serious company. The safe company. The company that understands the risks and should therefore be trusted with more customers, more capital, more government access, and more influence over regulation.
If a company sells AI to banks, law firms, governments, hospitals, and large enterprises, “responsible AI” becomes a sales advantage.
If a company raises huge amounts of money, “we understand the risks” helps calm investors.
If regulators are coming, publishing policy proposals helps shape the conversation early.
If the public starts blaming AI companies for job losses, Anthropic can say: “We warned everyone. We proposed solutions. We tried to be responsible.”
What I'm building
Delegate tasks. Get software.
Give Vroni a GitHub issue, bug report, spec, or rough idea. It reads the repo, plans the change, writes code, runs checks, and works toward a review-ready pull request.
Take a look at vroni.comI respect your privacy. Unsubscribe at any time.