Bernie's plan sucks, actually
TL;DR
The 50% seizure is the core problem: Shapiro says Bernie Sanders' AI sovereign wealth fund is effectively a one-time windfall tax paid in equity, giving the government controlling stakes and board power over labs like OpenAI, Anthropic, and xAI.
He likes sovereign wealth funds, just not this design: Shapiro is explicitly pro-sovereign-wealth-fund and ties it to his post-labor economics work, but argues real funds should be passive, broadly diversified, and constitutionally protected like Alaska's and Norway's.
Board seats and veto power are where the proposal turns dangerous: He calls democratic control through government voting shares and equal board representation a form of central planning that creates massive political capture and conflicts of interest.
Bernie is targeting the visible labs, not the real profit centers: Shapiro says frontier labs have huge cash burn and may not even be cash flow positive, while the actual winners are hyperscalers and infrastructure players like Microsoft, Google, Amazon, Nvidia, TSMC, ASML, and CoreWeave.
A workable version would tax the ecosystem, not confiscate specific companies: His alternative blueprint includes compute levies, windfall surtaxes of roughly 15% to 25% on monopoly-style AI rents, and public warrants tied to subsidies, plus a 10% cap on holdings based on Norway's model.
The payout should be automatic and visible to citizens: He sketches a template where 40% of returns become an AI dividend, 25% funds worker transition, 15% supports public infrastructure, 10% compensates local grids and communities hosting data centers, and 10% gets reinvested.
The Breakdown
A one-time 50% stock grab of OpenAI, Anthropic, and xAI is not just bad policy, David Shapiro argues. It's likely unconstitutional, invites capital flight, and misses the real AI money in Microsoft, Google, Amazon, Nvidia, and the broader compute stack.
Was This Useful?
Share
Keep Reading
Make Alcreon Yours
Tune your feedFive quick questions, and the feed ranks what matters to you first.Or just get notified
The weekly Echo. Signal worth keeping in your inbox.
Every new piece, announced on X.
Read Next
See all
Playbook
Tasteful Skills
“Tasteful Skills” argues that the best agent skills are not documentation or best-practice lists.

Playbook
The Art of Tasteful Prompting
Learn how tasteful prompting helps you move beyond generic AI output by shaping context, style, and judgment from the start.

Playbook
The Codex /goal Playbook
OpenAI shipped /goal for the Codex CLI. It turns a prompt into a persisted, self-continuing contract.