There is a tremendous potential with AI technology that goes beyond the limited scope of corporate models of profit and commodification. If this potential has any hope to be populist and progressive, it must also be an anti-capitalist approach to AI overall.
The following is a synthesis of various theoretical frameworks based on the work of Marshall McLuhan and David Graeber along with the ideas of universal basic ownership and the “direct from imagination era” that creates a coherent method for a truly democratic and anti-capitalist AI development. This synthesis reveals that AI can be an extension of human cognitive faculties and an opportunity to eliminate meaningless labor, while requiring new ownership structures that can foster a democratization of creative activity and a more participatory governance structure of the economy.
Marshall McLuhan wrote in Understanding Media: The Extensions of Man that technologies are extensions of human sensory and cognitive abilities that fundamentally reshape perception, social organization, and consciousness. In this framework, AI extends not just one faculty such as with print or radio. AI potentially extends all cognitive functions including memory, reasoning, creativity, judgment, and language itself. The real power of AI is not its specific outputs but its transformation of knowledge, labor, social relations, and power structures. A critical implication of this transformative potential is that if AI is an extension of humanity, then the question of who owns these extensions becomes an existential issue. Therefore, the status of AI must be outside of the control of private corporate interests. In other words, this technology has grown away from being just another commodity.
The core insight of David Graeber’s Bullshit Jobs is that a significant portion of modern employment consists of meaningless, unnecessary, or actively harmful work that even workers themselves recognize as pointless. This is not inefficiency but a political choice to maintain social control and distribute purchasing power in a financialized economy. The purpose of technology should be to eliminate unnecessary jobs, but capitalism creates more of them. AI has the ability to eliminate this type of job, but under current capitalist ownership it will likely create new categories of pointless labor.
These two concepts, technology as the extension of the human body and the potential of AI to replace unnecessary work, can be carried out through the concept of universal basic ownership. Universal basic ownership, developed by people such as Max Thake, Hilde Latour, and Trent McConaghy, is a shift from income redistribution to asset redistribution. Rather than giving people a basic income from tax revenue, universal basic ownership democratizes ownership of the productive assets that generate wealth. These assets include AI systems and automated technology. Users own the technology platforms they use, and profits from AI automation flow to asset owners in the form of ownership shares. Universal basic ownership argues that humans should own and profit from automated production rather than receive handouts while corporate elites control the machines. This public benefit requires the ultimate goal of democratic ownership of automation.
“The Direct From Imagination Era” is a concept developed by Jon Radoff that describes a paradigm shift where AI can potentially enable anyone to “speak worlds into existence.” This includes using natural language to create software, games, simulations, and virtual environments. The emphasis shifts from technical skills to the imagination. Individual users can speak or type what they want, and AI generates the implementation. There is a real-time instant feedback loop between imagination and manifestation. The result is a radical reduction in barriers to creating software, media, and experiences, and the potential for a massive expansion of creative participation. Due to the nature of the learning feedstock for AI systems, the type of work that is initially generated will be based on pastiche or simulation of preexisting styles in new applications. But over time, with experimentation of the way language is used to instruct AI, the form of creativity will evolve beyond the imitation of style. Therefore, it is important to understand who owns the AI platform and how to avoid the commodification of the products of the imagination.
In order to address the pressing current criticisms of AI while at the same time acknowledging the potential benefits of extending human cognitive functions, reducing unnecessary work, expanding ownership of AI systems, and equal access to the expression of the imagination, specific policy proposals must be made. These proposals may include:
- Training data should be collectively owned or licensed under peer production terms.
- AI infrastructure should be worker-owned, community-owned, or publicly owned.
- AI-generated wealth should fund social wealth funds with dividend distribution.
The machines doing work should always benefit the people rather than replace them. Ownership of these machines by economic elites is the primary factor in workers being replaced. Decentralization of ownership of AI systems prevents capture by these elites. Therefore, a broad-based ownership of AI and automated technologies is inherently populist. What this technology can potentially do to enhance human existence, once it has been removed from the capitalist structure, is inherently progressive.
The core principles of a populist and progressive form of AI emerges from these various perspectives. AI systems that extend human cognition should be collectively owned, no private entity should control the “extensions of man,” and individuals should own their cognitive data and its derivatives. AI should eliminate pointless work with an emphasis on the reduction of working hours rather than the creation of makework jobs, with social wealth distributed independently of employment status. Worker cooperative ownership, community ownership, and public ownership of AI systems must be paramount, with a decentralized infrastructure to prevent the concentration of power. There must be universal access to AI creation tools through these community owned platforms for the process of “imagination to manifestation”.
The theoretical frameworks of McLuhan’s media theory, Graeber’s “bullshit jobs” critique, Universal Basic Ownership, and the Direct From Imagination Era converge on a common insight. It is that the transformative potential of AI depends entirely on who owns and controls it. Since AI extends human cognition, those who control AI also control the extensions of our minds. Under capitalism, AI will create new forms of pointless labor. Therefore, only democratic ownership ensures liberation. Democratic ownership can be defined as broad ownership of AI assets, not just a redistribution of income. The goal is automation that ensures universal human ownership and benefit. The creative potential of imagination directly expressed through AI is also democratized. The short-term solutions of watermarks, peer production licenses, decentralized infrastructure, cooperative AI services, and social wealth funds are not isolated policies but components of a unified strategy to prevent corporate capture and build democratic alternatives. The long-term applications of creative assistants, personalized education, simulation tools, supply chain coordination, and democratic information systems are not speculative luxuries but achievable goals if we build the infrastructure and institutions now. Humans are not just managing a technology but shaping the future of intelligence itself. The central question is not whether AI will transform society, but whether that transformation will be democratic or authoritarian, liberatory or exploitative, inclusive or exclusive. This integrated framework provides the strategies and ethical foundations for choosing the former over the latter.
