OpenAI proposes four-day workweek to share AI productivity gains

April 9, 2026 OpenAI is proposing a four-day, 32-hour workweek as part of a broader policy plan to ensure workers benefit from AI-driven productivity gains. The company suggests running pilot programmes with no loss in pay, allowing reduced hours while maintaining output levels.

The proposal appears in a 13-page report titled Industrial Policy for the Intelligence Age: Ideas to Keep People First, which outlines how governments and companies can adapt as AI reduces the time required to complete routine and complex tasks.

OpenAI argues that AI systems are already shifting how work gets done. Tasks that once took hours can now be completed in significantly less time, and the company expects future systems to handle projects that currently take months. This, it says, will fundamentally change how organisations operate, how knowledge is created, and how work is structured.

To balance those gains, the report recommends that productivity improvements be shared more directly with workers. One approach is “benefits bonuses,” where employees receive a portion of efficiency gains through increased pay, financial security, or additional personal time.

The company also proposes the creation of a “public wealth fund” tied to AI-driven economic growth. The idea is to allow broader participation in the financial upside of AI by investing in companies building or deploying the technology, with returns distributed directly to citizens.

Together, the proposals reflect a shift in how AI’s economic impact is being framed. Rather than focusing only on efficiency and cost reduction, the report centres on how gains are distributed as workloads shrink and automation expands.

The recommendations come at a time when AI tools are moving beyond simple task support into handling more complex workflows. OpenAI’s position is that current policy frameworks may not be sufficient to manage that transition, particularly as the pace of change accelerates.

Top Stories

Related Articles

April 9, 2026 Farming equipment manufacturer John Deere has agreed to a $99 million settlement in a long-running right-to-repair dispute more...

April 9, 2026 Residents in Toronto’s affluent Rosedale neighbourhood are considering an AI-powered surveillance system to create what would be more...

April 9, 2026 Kyndryl has introduced a new Agentic Service Management offering designed to help enterprises transition from traditional IT more...

April 8, 2026 Developers are raising concerns that Anthropic’s Claude Code is becoming less reliable for complex engineering tasks, based more...

Jim Love

Jim is an author and podcast host with over 40 years in technology.

Share:
Facebook
Twitter
LinkedIn