What makes a human a human? Can AI think like one? Is it going to one day become smarter than us? And most importantly, did Karl Marx consider this back in the 1800s? Well, okay, the answer to that is yes, let’s just say that now. Pretty much anything you could ask if Marx considered, he did.
This article contains an analysis of artificial intelligence from a dialectical materialist perspective, and a short guide on how you can move away from the capitalist monopoly on tech and toward a democratically owned future for cyberspace.
Since the dawn of time, humans have adapted old tools and created new ones in order to do two things: overcome the limits of humanity and accumulate wealth. However, generative AI, as it currently exists in society, is less of a tool and more a weapon of the bourgeoisie in the modern day workers’ struggle against the capitalist relations of production.
The “Quasi-Human” Bourgeois Lie
LLMs as they currently exist are powerful, but they are isolated from material reality. Their outputs, while they may be somewhat coherent, often lack practical truth when closely scrutinized. Any claims of “AI consciousness” from the tech giants developing these tools are simply idealistic ones; they are not grounded in any sort of objective reality.
AI is merely an imitation of the structure of the human brain, not an exact replica. It will not and cannot reach the peak of human thinking, let alone surpass it. This idea of a future where humans will no longer need to work and machines will supposedly be what liberates them is nothing more than a utopian ideal that ignores the overwhelming amount of problems with the capitalist system itself.
When we look at the process of production, it is clear that AI does not meet the threshold for humanity. The producer is the subject, the dominant force, while the means of production are either intermediaries or passive objects. Artificial intelligence qualitatively falls under the means of production, as its development can only take place with a producer behind it. It can never break through the threshold of this degree to change into a producer of its own. Put simply: if humans pull the plug on AI, it is over. This is not a sci-fi movie, there is no evidence to support the idea that AI models could somehow find ways to avoid being shut off.
Furthermore, human wisdom can replicate the full historical evolution of human thought, examining thought as a subject of contemplation in of itself. This allows for the simultaneous viewing of social processes as historical ones, leading to a more acute understanding of social responsibility and one’s historical purpose. Social and historical unification in this way is impossible of a machine because the individual is a social being and the human understanding of anything at any given stage is non-supreme, infinite and relative.
Artificial intelligence can never develop into a human being, and it is not ‘quasi-human’ in any real sense. Knowledge will remain inexhaustible for humanity, and a perfect, absolute simulation will never be attained. Thus, AI’s capabilities will always remain below human intelligence and there will never be any “superhuman” advancement of this technology, no matter what the bourgeoisie CEOs claim.
The Economic Struggle of Late-Stage Capitalism
What we have named scientific logic is, in essence, a component of dialectical thought. Through dialectics, we can understand that reality is a continuously transforming process, shaped by internal oppositions and its own negation. From this dialectical viewpoint, AI, mirroring other scientific progress, represents social evolution for humanity while also serving as a tool for capital.
Under capitalism, science functions as a means to generate profit. It facilitates the continuation of capital by exploiting workers, rather than pursuing the advancement of humankind. This has been proven by history before, during England’s Industrial Revolution, and Marx’s critique of political economy explained how this tendency is self-destructive, as capital’s survival mechanisms are inherently linked to its eventual destruction.
Under capitalism, the technological alienation of labor leads to the domination of machines over workers in the struggle between workers and machines. Creativity, previously thought to be uniquely human and superior to the imitative abilities of other species, is now being supplanted by machines. Leading bourgeois institutions in applied science, including Microsoft, Google, Meta, Alibaba, and OpenAI, regularly release their language models and frameworks aimed at automating creative and intellectual pursuits like writing, design, translation, and music creation.
In contrast, numerous countries, many of which are countries that are being exploited by the imperialist nature of capitalist nations like the USA, still rely on human labor for degrading and dangerous jobs such as mining, construction, etc. Evidently, technological advancement under capitalism only serves the interests of the ruling class. Instead of freeing people from this dangerous work, AI is being used to increase feelings of isolation, disconnect workers from their innate creativity, and turn them into mere tools for mass production and duplication.
As capitalism faces a crisis of overproduction and falling profits, the rise of AI is proof that mere technological advances are insufficient to rescue the system. The tendency for the rate of profit to fall occurs when the “constant” composition of capital increases, signifying a rise in investment in machinery and automation over living labor. As human labor is the sole source of surplus value, it consequently decreases. To counteract this declining profit rate, capitalists must cut wages, either by lowering individual pay or by reducing the workforce – both a common occurrence today.
The bourgeoisie and its worshippers perpetuate the myth that AI will generate new, well-compensated employment, but reducing variable capital (wages) to the bare minimum for survival is the system’s objective. A rise in constant capital would speed up the profit rate’s decline, which is exactly what capital tries to prevent, leading to widespread job cuts in various sectors. Therefore, it is undeniable that capitalism prevents AI from freeing any working class human. It embodies capital’s base conflict: constantly pushing for more production yet weakening the source of surplus value – human work. For technology and art to truly liberate and express humanity without alienation, the means of production, including AI, must be released from private ownership and aligned with social needs over profit.
Intellectual Property as a Branch of the Bourgeoisie
From a Marxist perspective, we should diverge from the liberal critiques of generative AI, such as those concerning copyright infringement. The issue goes beyond the mere “theft” of intellectual property, as the very idea of “theft” rests on the legitimacy and sanctity of private ownership (different from personal property), which we must reject entirely.
The historical context of capitalism defines the concept of private ownership, rather than it being a permanent requirement. “Theft,” as defined by neoliberalism, is a re-establishment of capitalist relationships. AI’s actions of “copying” or “appropriating” do not violate ownership; instead, they apply the principles of private property to the realms of thought and creation.
Marx asserted that the true problem lay with ownership, not the act of stealing it, and he was right to. Modern economies thrive on information control and processing, which has led to information access and data control becoming a formidable new source of power and tool for oppression. The contemporary bourgeoisie, the corporate elites, are systematically accumulating ownership of information and concepts. Meanwhile, intellectual laborers, those who create and operate within the sphere of information and concepts, face escalating exploitation, alienation, and are denied the true worth of their work’s output.
This concentration of property and exploitation of intellectual labor and its products into the hands of a few is facilitated by the structure of intellectual property rights. Law, as a state-sanctioned system of enforceable rules, mirrors the primary mode of production and its associated social relationships. It functions to ensure the perpetuation of these relationships and, by extension, to solidify and advance the interests of the ruling class through the continuation of its authority.
Historically, we can see this exploitation through the routine denial of IP rights for Black, Indigenous, and other marginalized peoples who sought out protections for their ideas and creativity. We must, of course, oppose big tech exploiting artists and creativity under these material conditions, but our opposition must remain dialectical. The criticism here should not be about “IP theft”, but rather about the uneven production relationships enabling such appropriation.
To sum all of that up in one sentence: Corporate control over knowledge, culture, and technological functionality is established through the sham of intellectual property like copyrights and patents, which enables them to boost user engagement, monopolize innovation, and profit from data that should be democratically accessible and owned. According to economist Dean Baker, consumers overpay by approximately $1 trillion each year due to intellectual property price hikes, which he believes could be avoided in a “free market” without patents or copyright monopolies. Replacing intellectual property with a commons-based knowledge sharing model would lower costs, broaden educational access, and serve as a mechanism for wealth redistribution and reparations to the Global South.
The Ecological Trade-Off of Innovation
To ignore the fact that AI accelerates the social-ecological catastrophe, the climate catastrophe in particular, is also naive. This is a frequently accepted trade-off by neoliberals and conservatives alike, with many arguing that not all AI is bad, and that it can “be a helpful tool” for humanity. But is sacrificing science and our ecosystems really worth it for technology that hasn’t been perfected yet?
The spread of AI weakens community connections and biodiversity, leading to a decline in social and ecological intelligence. These publicly-accessible chatbots may offer people more instant knowledge than ever before, but they limit real scientific exploration, foster research feedback loops, and have severe impacts on the environment, which is already under immense attack. Researchers have pinpointed nine worldwide metrics for ecological sustainability. Seven of them are estimated to have reached their danger limits. Capitalism’s inherent need for accumulation has caused us to surpass at least six planetary boundaries, such as climate change, ecosystem integrity, nutrient cycles, water resources, land use, and chemical pollution.
This destruction primarily harms the poor, particularly in impoverished nations, and the harshest effects of capitalist issues fall on marginalized and racialized communities. They are deliberately settled in hazardous waste sites and underdeveloped areas. These populations, affected by environmental racism, are also systematically excluded from the design and implementation of environmental policies. The “need” for data centers is rapidly increasing the plundering and colonization of land globally.
In 2023, US data centers used roughly 211 billion liters of water, with this number projected to more than double by 2028. In 2024, 8,000 data centers worldwide used 460 TWh of electricity annually. Now, in 2026, this figure is projected to increase by an additional 160 to 590 TWh, comparable to the yearly energy usage of Sweden and Germany, respectively. According to the International Energy Agency, these infrastructures’ CO2 emissions are set to triple between now and 2035.
Annual waste from AI’s rare earth extraction is 13 billion tonnes globally, and some forecasts predict it could exceed that by over a hundredfold by 2050. Once again, these effects disproportionately impact the impoverished globally, either through direct consequences like mining and water depletion for data centers, or indirect ones such as biodiversity loss and severe climate events. While it is true that AI technology can have helpful applications in fields such as health, science, etc – all of those applications are currently prevented from truly benefiting the working class due to capitalism’s inherent need for the devaluation of the worker and the maximization of profit.
There is also a clear difference between the technology that is being worked on by big tech and the technology that would actually accomplish the benefits they are claiming are possible. The algorithms helping with things such as data analysis, scientific pattern recognition, or flood prediction are not the same type of AI as a consumer-facing chatbot. AI companies and fanatics are pushing the idea that these huge, resource-hungry models are indispensable, yet many smaller, more efficient options are available at a significantly lower cost, for both individuals and the planet.
Resist Capitalist Capitulation, Choose Cyber-Socialism
At the end of the day, there’s no real evidence for AI ushering in a new golden age. These tech billionaires are lying to you, because they do not want their investments wasted. Estimates suggest that AI adoption will lead to annual productivity growth of 1.8% annually over the next ten years. This is far from enough to sustain any sort of prolonged growth – and keep in mind that it is primarily the AI companies themselves making these estimates. Evidently, capitalism cannot escape the inherent problems of value production through technological advancement alone.
Generative AI also possesses a horrifying ability to sway people’s opinions. We’ve seen this with the tragic cases of young children committing suicide, and people convinced to murder their loved ones due to talking to these chat bots. Creating images and texts has always been a powerful way to indoctrinate, triggering “rigid thinking” in the brain that is effectively a form of trauma. According to neuroscience researchers, these trauma mechanisms may result in epigenetic modifications passed down through several generations. If accurate, corporate-owned, unregulated generative AI usage has the potential to permanently re-enslave humanity to the irrational beliefs we outgrew due to scientific progression.
None of this has to be this way. We can still change the trajectory we’re on, and we have all of the power, despite what the bourgeoisie wants us to believe. Contrary to what you may have heard, neither socialists nor communists are opposed to technological advancement, and neither socialists nor communists are going to make you share a toothbrush.
Under socialism, assets are collectively owned. Workers themselves directly manage the means of production via worker co-ops, with output geared towards meeting needs and use, not for exchange, profit, or accumulation. There are differing opinions among socialists/communists regarding the state’s role. Some favor highly decentralized governance and economic production, while others believe in more extensive state planning. The digital economy should utilize these same principles, strategies, and tactics. The cyber-socialist model proposes ending intellectual property, socializing computational infrastructure, democratizing data and digital intelligence, and entrusting the development and upkeep of the digital ecosystem to public communities. What’s even better is that the foundational elements for a socialist digital economy are already largely in place.
A socialist mode of production is enabled, for example, by the software and licensing provided by Free and Open Source Software (FOSS) and Creative Commons licenses. There are also plenty of co-ops in general that are successful even while working within the limits of a capitalist society. Through open-source public interest tools like DECODE, cities in the EU can empower citizens to engage in community activities, providing access to and contribution of data on matters such as air pollution and online petitions, while ensuring the individuals retain ownership of their shared data. The Fediverse, a collection of interconnected social networks using shared protocols, offers a socialist social media alternative that promotes decentralized online communication.
Here at The Convergence Lens, we are also a co-op that is democratically managed. Both this platform and our production company, Convergence Lens Studios, are run by the people, for the people. We use as much open-source software as we can to run our site, and we hope that our community platform can help contribute a little to the socialist approach to cyberspace, particularly among activists, advocates, and community organizers.
But we, along with all of the people fighting this fight alongside us, need as much support as we can get. Choose to free yourself from the shackles of capitalist corporations by supporting open-source software developers, worker co-ops across all fields, and creatives that share their work independently. Support regulations and reforms that bolster individual privacy safeguards, curb state surveillance, dismantle the copyright empire, and abolish private profit carceral systems globally.
A Beginner’s Guide to Cyber-Socialism
There will need to be massive governmental restructuring across the globe in order to truly develop an eco-socialist focused internet. Even though they are presented as progressive, environmental proposals like the US and European Green Deals are still rooted in capitalism and perpetuate its negative aspects, including endless growth, imperialism, and systemic inequality. Better alternatives are presented by eco-socialist, indigenous-led models like the Red Nation’s Red Deal, the Cochabamaba Agreement, and South Africa’s Climate Justice Charter. These plans accept that growth has limits and include rules for moving to a sustainable economy. Aside from pushing for these types of models to be enacted in your city, state, country, whatever it may be, here are some ways you can begin to transition away from Big Tech and towards a cyber-socialist future:
- Stop Using ChatGPT: As QuitGPT’s website says, ChatGPT/OpenAI is Trump’s biggest donor, and ICE along with other federal agencies partner with ChatGPT to automate systems used to hunt down minorities. If you absolutely need generative AI (ask yourself if you do or if there are other tools that may help you in a less harmful manner), open-source alternatives, such as the ones found on Open Alternative’s website, are your best option. Using self-hosted tools and open-source models instead of corporate ones, further helps reduce the need for data centers. However, we understand that learning how to deploy and manage self-hosted tools can be hard. At the very least, switching from ChatGPT to Claude or Gemini reduces some degree of harm – but these are not radical options by any means, they are still capitalist corporate LLMs. Going a step further, you can use privacy-focused alternatives such as Confer or Lumo. But again, self-hosting is always going to be the best option if you can learn how. This Self-Hosting Guide is a great starting point. China’s ByteDance recently dropped a self-hosted desktop agent that is still in beta, but offers the potential to move away from USA-based models. It is a potential alternative to Claude Code/Cowork. “A good model works well and is affordable to everyone,” Tan Dai, president of Volcano Engine, ByteDance’s cloud computing services unit, said in 2024.
- Build a Home Data Server: This option is not possible for everyone, and that’s understandable, but if you can, switching to a home server for things like your photo and video storage, website hosting, email, chats, etc, not only protects you from being surveilled by Big Tech and governments, but reduces the need for these corporations and their harmful effects. FreedomBox provides free, open-source software for affordable personal servers. These servers can host and route data for various services, including email, calendars, chat applications, social media, and other functions. You can build your own hardware and deploy the open-source software or buy their premade boxes. The AS5402T by Austor is another NAS that offers an easy way to get started due to the amount of apps that come built in.
- Secure Your Data: Even if you cannot set up a whole storage system, there are ways to secure your data from your laptop or computer. Projects like Solid give you the ability to keep your data in self-controlled “pods.” Users maintain control over all of their data, allowing app providers, social media networks, and other services to access it only under user-approved terms. Cryptomator can help lock down your hard drive to prevent unwanted access of your files.
- Focus on Open-Source Software and Mutual Aid: Rather than supporting businesses that aid the capitalist means of production, support open-source software and mutual aid networks. Open Alternative is a good site to find things you need, as mentioned earlier, but there is also Alternative To. You can use Mutual Aid Hub to find and submit mutual aid groups near you, and you can visit Mutual Aid Co-Op to find tools and resources for how to set up mutual aid networks in your own community. The Mutual Aid Network is a global cooperative. Business owners, project managers, and individuals with ideas can join to get community support for finding and succeeding in work they’re passionate about.
- Support Community Ownership of Physical Infrastructure: Any development of cloud server farms, cell towers, fiber optic networks, and transoceanic submarine cables should be led by the people. By supporting community-run internet service providers and wireless mesh networks, we can put these services directly under community control. An international consortium could manage certain infrastructure, like submarine cables and satellite data, by building and maintaining them affordably for public benefit instead of for profit.
A Community-First Future
All of the above is no doubt easier said than done, but none of it is impossible. We have the power to change things, and we have a strong foundation for how we should be structuring these systems through work that is already being done. You can help by taking the steps to begin de-centering corporations and capitalism from your own life as much as possible. Connect with your community, find those mutual aid networks, and help build the infrastructure needed to create a society that truly prioritizes people over profit.
