The Byte is The AI Collective’s leading publication for thought leaders sharing about the incredible discoveries they’ve made and projects they are working on.

For more information or how to get in touch with The Byte editorial team, send us a message at [email protected].

~ Josh Evans, Managing Editor

The debate over how to respond to the economic impacts of AI is littered with false binaries. Is the best solution stronger safety nets, or a restructuring of our economic foundations? Should we focus on retraining workers or a universal basic income? These questions presume we must choose – that one approach must be right and others wrong.

These framings miss the essential point: We need distinct interventions to manage the different stages of the transition to an AI-driven economy. Right now, the best move for those of us exploring these topics is to be prepared - to visualise and describe the entire roadmap from near-term disruption to the inevitable restructuring of the economy. With an integrated roadmap, the wide range of policy ideas can become more coherent, and we can better help policymakers and entrepreneurs look further down the line into a very different future. Below, we outline one such roadmap.

Economic forecasts of AI vary heavily between AI futurists and establishment thinkers. On one side stand technologists who predict unimaginable growth and warn that automation will eliminate most jobs. On the other hand, traditional economists such as Nobel Laureate Aghion predict outcomes more on the order of 1% annual productivity gains, viewing AI as a normal technology that will create more jobs than it destroys.

We believe that there is a useful framing that spans these perspectives – that all of these people are describing the same roadmap towards economic prosperity, just with very different timelines. Anton Leicht has a strong piece integrating these framings, discussing two upcoming phases of automation.

Our most important task during this period is to prepare and to guide society through these sequential transitions. To do so, we must develop thoughtful roadmaps that account for both the near and long-term impacts and that can adapt effectively to the progressively increasing demands on our governments to manage these changes.

In the near term, we are primarily concerned with AI economic shocks, recessions, and the labour displacement of certain groups, such as early-career employees. There will certainly be other jobs to transition into. The only question is whether they will be accessible and desirable. History suggests that entire regions could lose their advantages. Without good jobs to transition into, we could see the same shocks that trade brought to the American Midwest, where regions of economic impoverishment have persisted for decades. Recent research by AI economists describes similarly concentrated pockets of potential vulnerability in the U.S. labor market.

The lack of a clear path to upward mobility for entry-level workers may become a pressing concern. Even with less desirable opportunities available, this could generate significant disillusionment among a generation that invested heavily in education, expecting traditional career trajectories. Reskilling is a valid and crucial response – though what jobs to reskill for remains an open question when the target itself keeps moving.

For entrepreneurs, this volatility creates distinctive opportunities. One-person unicorns become possible as AI amplifies individual productivity to unprecedented levels. While young people find it increasingly hard to secure entry-level roles at established companies, more will deploy AI for their own purposes, launching ventures that previously required substantial teams and significant capital. Startups can compete with incumbents on execution speed, compressing timelines from years into months. However, finding experienced talent to guide AI workflows becomes the critical bottleneck. Domain expertise and seasoned judgment become increasingly valuable precisely as routine execution is automated. The founders who thrive will be those who can orchestrate AI capabilities.

In the medium term, the transition to an AI-dominated economy will present both extraordinary opportunities and structural risks. AI systems will be able to complete ever more workstreams end-to-end, and improving capabilities will aid a new class of superstar firms in winner-takes-all markets where scale – especially of data and compute – confers decisive advantages. This period could be defined by a delayed but rapidly accelerating impact on production and employment, the erosion of traditional pathways to upward mobility, and growing pressure on fiscal systems.

At the same time, some risks warrant attention. Founders that don’t scale quickly need to sell or can’t compete with AI-first competition. This will contribute to a loss of upward social mobility and lead to a growing divide between capital holders on one side and startups, small businesses, and workers on the other. Meanwhile, increasing automation will test whether governments can adapt institutions fast enough to manage structural displacement.

In the long term, the most likely trajectory of AI technology, provided we avoid larger risks, is that it eventually surpasses human intelligence by orders of magnitude. Machine intelligence faces fewer fundamental constraints than biological cognition. AI will continue to decrease in cost, both for intellectual computation, which already appears to be an order of magnitude cheaper than human labor, and for manual labor, constrained primarily by the marginal cost of robotics. The end state is almost certainly the automation of substantial value creation, with durable human advantages persisting primarily in domains requiring human connection, physical presence, and contexts where people specifically prefer human involvement.

This transformation will challenge the foundational premise of modern economies: that labor is the primary mechanism through which individuals access income, status, and economic security. As labor ceases to be a reliable path to capital accumulation, core aspects of the social contract (e.g. “if I work hard, I can get ahead”) will break down for a growing share of the population.

We will likely engage with ideas of universal basic income (UBI), reduced work-week expectations, and new institutions that provide meaning and community beyond traditional employment. A particular proposal that preserves human agency is universal basic capital (UBC), which distributes the windfall from AI fairly by giving everyone at birth a stake in the economy. Unlike UBI, UBC addresses inequality by granting citizens permanent property rights rather than providing recurring transfers. UBC would maintain citizens’ status as stakeholders rather than as dependents on transfers, thereby making democracy more viable.

For entrepreneurs building today, this trajectory suggests two strategic orientations. First, business models built on labor arbitrage will collapse when AI labor costs approach zero: Any venture premised on doing something cheaper with humans has a limited shelf life and is transitional rather than durable. Second, competitive moats shift back toward large businesses as they gain superior access to intelligence through compute resources and proprietary data. The window for startup advantage may be this current transitional period, when agility and speed matter more than scale and resources.

Source: Korinek and Suh (2024)

At some point, perhaps not as near as futurists think, but nearer than establishment thinkers believe, we will be forced to grapple with these questions collectively. Transformative AI will challenge our fundamental assumptions about how economies function. Each stage of this transition will demand a different set of tools.

In a coherent roadmap, each layer of solutions serves as a prerequisite for the next stage. Near-term payment rails would eventually become the delivery mechanism for benefits reform.

Fundamentally, we all share a common purpose: guiding our society towards a future with abundance and shared prosperity for the many, not the few.

The Byte is The AI Collective’s insight series highlighting non-obvious AI trends and the people uncovering them, curated by Noah Frank and Josh Evans. Questions or pitches: [email protected].

Roam is the virtual workspace our team relies on to stay connected across time zones. It makes collaboration feel natural with shared spaces, private rooms, and built-in AI tools.

Roam’s focus on human-centered collaboration is why they’re our Premier Partner, supporting our mission to connect the builders and leaders shaping the future of AI.

Launching a new product or hosting an event? Put your work in front of our global audience of builders, founders, and operators — we feature select products and announcements that offer real value to our readers.

The AI Collective is driven by our team of passionate volunteers. All proceeds fund future programs for the benefit of our members.

Are you subscribed across our platforms? Stay close to the community:

About Deric Cheng

Deric Cheng is Director of Research at Windfall Trust, where he focuses on how societies can navigate the economic transition driven by advanced AI and ensure the gains are broadly shared. He also leads the AGI Social Contract, a consortium developing concrete policy ideas for a post-AGI economy, and has convened economists and policy experts through efforts like Convergence Analysis’ Threshold 2030 conference and related fellowships. Before shifting into AI economics and governance, he was an early engineer at Alchemy (where he launched Web3 University for 200k+ learners) and a prototyping researcher at Google’s Interaction Lab, where he helped build early next-gen Google Glass explorations and the first real-time translation feature for Pixel Buds.

About Jacob Schaal

Jacob Schaal is an economist studying how AI reshapes work at the task level, with a focus on who gets hit first, especially early-career professionals. He is a Research Manager at ERA Cambridge, where he coordinates AI governance research and built an automation exposure index grounded in Moravec’s Paradox, and a Research Assistant at King’s College London researching AI’s labor market effects. He also edits the AI Economics Brief at Windfall Trust, read by over 1,200 people across frontier AI labs, government teams, and academia, and has worked on EU AI Act implementation and international AI standards and due diligence work in Brussels.