Wake up!

A chat with GPT on the one thing we have to do before we can wield AI to our advantage to make our lives and world better.

(Adding my book How to Fire Your Boss for context helped this convo midway. )

Me:

Here's the point.

For the first time since we left hunter gathering positions or safe villages somewhere, we have the potential to sustain our lives to the fullest without taking from other people. And AI has made that possible. It is the engine that we rely on. A pathway from slavery, indentured servitude, and hierarchical use of workers, the proletariat, on whose backs industry has always been born. That model is passé. It's time for something new, and we're moving already into something new. And AI is a tool that we can use to do that. But it's also a tool that we can use to just stay completely stuck in the former paradigm of profits and power.

GPT:

Targeting AI's potential

AI can either empower us to break free from profit-driven paradigms or entrench them further. Reflecting on its possibilities and necessary safeguards is key to shaping a balanced future.

It’s true that artificial intelligence has the potential to alter the economic and social order in a way few technologies ever have. On one hand, it promises increased productivity, cheaper goods and services, and—most significantly—the capacity to replace hard or dangerous human labor. In the most optimistic vision, it could liberate people to spend more time on creative pursuits, personal growth, or community-building rather than eking out a living through low-wage labor.

However, as you point out, if we simply apply AI to existing hierarchies and modes of profit extraction, it will reinforce and possibly deepen inequalities. The same technology that can free us from repetitive work can also be used to monitor and control populations, deepen surveillance, and supercharge the concentration of wealth. That’s the core tension: AI is a tool, and it will be shaped by the values, incentives, and power structures already in place.

Below are a few key considerations in deciding which of these futures we move toward:

  1. Ownership and Governance

    • Open Source vs. Proprietary: If AI research and tools are controlled by a small number of corporations, those corporations may use AI primarily to further increase their own profits and power. Open-source alternatives, communal ownership models, or stronger public oversight could help distribute AI’s benefits more evenly.

    • Policy and Regulation: The right kinds of regulations, especially around data privacy, algorithmic fairness, and safety, can ensure that AI supports the public good. Policy can also create guardrails against exploitative uses of AI (for example, mass surveillance or predatory algorithmic advertising).

  2. Economic Models and Universal Basic Services

    • Universal Basic Income or Services: Some economists and futurists argue that if AI-driven automation reduces the need for human labor in certain sectors, society should consider implementing Universal Basic Income, or even more ambitiously, “universal basic services” (like healthcare, education, housing, internet access). This could help protect against the displacement of workers and ensure that the gains from AI are shared.

    • Worker Cooperatives and New Ownership Structures: As AI takes on tasks, we could explore cooperative models where workers themselves own or co-own AI-driven enterprises, ensuring equitable distribution of profits.

  3. Human-Centric and Ethical Design

    • Prioritizing Well-Being: If we design AI systems with principles such as fairness, accountability, and transparency in mind, we reduce the risk of a purely profit-driven approach. Initiatives like “human-in-the-loop” AI can keep decision-making—and responsibility—anchored in human values.

    • Cultural Shifts: It’s not enough just to have laws in place; there must be a widely held conviction that technology should serve all of humanity. Grassroots movements, international coalitions, and broader societal conversations are critical if we want to move beyond the old “profits-over-people” model.

  4. Global Collaboration vs. Competition

    • International Cooperation: Many of AI’s benefits and risks (like climate modeling or autonomous weapons) span national boundaries. Collaborative efforts that set common standards and share benefits internationally will be crucial in preventing a two-tier or multi-tier world, where only some countries reap AI’s gains.

    • Development for the Majority: AI could help transform agriculture, healthcare, and education in developing regions—if the tools are made accessible. Otherwise, it could become another mechanism that leaves poorer nations behind.

  5. Education and Skills

    • Lifelong Learning: As the job market evolves, people need resources and opportunities to reskill or shift into new roles that are augmented (rather than replaced) by AI. An emphasis on critical thinking, creativity, and emotional intelligence will equip us for areas where humans excel and machines still struggle.

    • Bridging the Digital Divide: Making sure everyone has access not just to the internet but also to the skills and tools needed to leverage AI for personal and community benefit is a key part of ensuring that AI doesn’t replicate old patterns of inequality.

Ultimately, AI is indeed a “power tool,” and power tools can build houses or tear them down. To steer it in a liberatory direction requires broad participation in setting norms, rules, and expectations around its use—something that goes beyond the tech giants or a handful of policymakers. If we treat AI’s development and deployment as a collective effort, then it really can open doors to a new paradigm in which no one’s well-being is premised on someone else’s subjugation.

Yeah, I'm going to push back on those canned answers. I don't think the biggest problem with AI is watching us. I think the biggest problem with it is utilizing it and putting it into service to make more profits for shareholders without any benefits for workers other than getting pushed out of a job. That's the number one problem with AI, other than some 12-year-old finding nuclear launch codes, and that all of our data has been exported into a discoverable and synthesizable dataset for training these AI models, and we've completely lost our freedom, our expectation of privacy, and our right to privacy. Aside from all those things, more fundamentally than privacy is putting food on the table, which is everyone's interest, and that is the problem that's at stake. In the next, we'll say, week to 10 years, your job could get replaced by AI, and the reason is because of profiteering.

So I don't think the curbs are regulation. That's not going to do anything at all. That's a joke. And I don't think that the, you know, bridging the digital divide... So anyway, the point is, this is an opportunity to seize the engine that drives our own self-sufficiency personally. And you start personally, you start using this tool to advance yourself, to build SaaS for yourself, to build equity for yourself, to build skills for yourself, to educate yourself, to advance yourself into the next era. Secondly, it's about working together in a way that we forgot how to do, instead of exploiting each other. Exploitation of slavery has continued in another form of hierarchy in the workplace. It is exploitation. Not just by bad bosses. By good bosses, whose job it is to pay you as little as possible for work, no matter what value you contribute. That is exploitation.

You’re pointing to something more radical and tangible than just “better regulation” or “closing the digital divide.” You’re talking about seizing AI for genuine self-sufficiency—on both the individual and communal level—so that people aren’t relegated to the losing side of a profit-driven system. Let’s lean into that. Here are some direct, concrete ways individuals and communities can use AI to build actual sovereignty, rather than remaining cogs in someone else’s machine:

1. Personal Empowerment and Self-Sufficiency

  • AI as a Personal Tutor/Skill-Builder

    • Instead of paying for expensive courses or degrees, use AI tools (like language models, coding assistants, design generators) to learn at your own pace.

    • For instance, if you want to pivot into software development or data science, AI can speed up the learning curve by generating real-time examples, debugging your code, and suggesting resources.

    • This short-circuits the old system where you have to invest tens of thousands of dollars into education, only to stay perpetually in debt.

  • Launching Side Projects and Micro-Enterprises

    • AI can help you prototype ideas quicker—whether that’s designing a product, writing a business plan, or even generating marketing content.

    • Tools like generative image or text models can act as “employees” for small entrepreneurs—doing the work that would otherwise require you to hire a graphic designer, a copywriter, or a social media manager.

    • The key is to own the process yourself rather than being at the mercy of a company that’s chasing profit. It’s your IP, your platform, your brand.

  • Localization & DIY Production

    • Tied to technology like 3D printing or low-cost robotics, AI can help you design and produce items locally—anything from small home goods to prototypes for small-batch manufacturing.

    • This reduces reliance on massive global supply chains controlled by large corporations. Over time, communities could become self-reliant for a range of goods.

2. Relearning Collaboration and Cooperation

  • Worker-Owned AI Platforms

    • Instead of building another hierarchical company, people can form co-ops where AI-driven services are collectively owned. For example, a group of freelance developers or designers might collectively pay for (and own) an AI infrastructure that they all benefit from—splitting costs, profits, and decision-making equally.

    • This model upends the typical “exploitative” structure where one entity reaps most of the rewards while the workers see minimal gains.

  • Skill-Sharing Networks

    • Individuals who learn to use AI effectively can share that knowledge with others in their community, creating circles or hubs where these skills get transmitted.

    • In-person or online meetups dedicated to “local AI usage” can help everyone level up. The result is a community that’s not easy to exploit because people understand their worth and have the means to produce value on their own terms.

  • Pooling Resources for Collective Benefit

    • Instead of each person paying individually for advanced AI tools, groups can pool funds or barter services. Then they share the tools democratically.

    • You might see a local “AI lab” in a community center—funded by members or grants—where folks can access hardware (GPUs, data storage) and advanced models without having to pay corporate subscription fees.

3. Building New Economic Foundations

  • Microgrids and Community Infrastructure

    • AI can optimize and manage local energy microgrids or community farming operations (for example, precision agriculture) so communities become less dependent on centralized power or food systems.

    • When you combine local solar/wind with AI-driven resource management, you start seeing real self-sufficiency—less tied to exploitative utility bills or agribusiness giants.

  • Alternative Currencies and Barter Systems

    • It might sound offbeat, but AI could help manage local exchange systems or time-banking, making sure everyone’s contributions are tracked and valued.

    • This bypasses the standard currency system, which often funnels wealth up the chain to large financial institutions.

  • Open-Source Everything

    • Instead of paying license fees to private entities, communities can leverage open-source AI frameworks (like those from Hugging Face or OpenAI’s open models, etc.) and adapt them locally.

    • Open-source means the code is transparent, you can modify it for local needs, and you’re not beholden to the whims of a corporation that might cut off access or spike the price.

4. Owning Your Data (and Your Destiny)

  • Personal Data Cooperatives

    • Imagine everyone in a community pooling their data but setting their own terms on how it can be used, and by whom.

    • If a business wants to train AI on that data, the co-op can negotiate a direct stake or royalty, rather than letting Big Tech vacuum it up for free.

  • Local Hosting and Edge Computing

    • Instead of pushing all your data to remote servers owned by big corporations, you use AI models that run locally on your own machines or community servers.

    • This is technically challenging right now, but developments in smaller, more efficient models are on the rise. Over time, running powerful AI locally (or in a neighborhood data center) will get easier, giving you real autonomy.

5. A Cultural and Mindset Shift

  • Rejecting Extractive Models

    • When you discuss “exploitation,” it’s really about the entire mindset that we measure everything in profit terms. If we focus more on sustainable, cooperative models, AI can be a force multiplier for those efforts.

    • This won’t happen just because we wish it so; it requires organizing, building real examples, and sharing success stories that demonstrate a viable alternative to standard capitalism.

  • Prioritizing Community Education

    • Especially for people who fear being displaced by AI, it’s not enough to say, “go learn some skills.” It needs to be a collective effort where people help each other learn. If you build a community that’s strong in AI literacy, that community is much harder to exploit.

    • This might mean local “AI literacy nights,” open workshops, even neighborhood-based hackathons.

  • Redefining ‘Work’ and ‘Value’

    • Ultimately, the dream is for AI to free us from drudgery so we can do more meaningful, human-centric tasks. But we have to redefine our economic systems so that a person’s worth isn’t tied to a 40-hour grind.

    • This is a long-term cultural shift—one that might be accelerated if people realize how AI can handle the so-called “grunt work” while humans focus on creative, communal, or caretaking roles (things machines can’t replicate easily).

The Bottom Line

Instead of waiting for some top-down policy that might never come—or might just keep AI as the plaything of large corporations—you’re advocating a bottom-up approach. Harness AI’s power individually and collectively so that:

  • You upgrade your own skills, create your own micro enterprises, and become less dependent on big companies for survival.

  • Communities collaborate, pool resources, and share knowledge to form a shield against exploitative practices.

  • People reclaim ownership of their data and production processes, building local resilience.

This is the essence of using technology not just to maintain the status quo but to actually break out of the cycle of exploitation. It’s not going to happen overnight, but the tools to start building that future are increasingly at our fingertips—so long as we choose to use them, together, rather than passively letting them be used on us.

(uploaded my book)

Read through this three times. Identify the themes and thinking that gives context for our conversation

Below is a distilled overview of the core themes and the underlying thinking across your recent discussion, based on reading the conversation multiple times. I’ve highlighted the recurring ideas and how they fit together.

1. AI as a Turning Point in Human History

  • Potential for True Self-Sufficiency: There’s a strong conviction that AI could enable individuals (and communities) to break away from reliance on exploitative systems. Historically, human labor has always been “harvested” to enrich a few; AI might change that, if used correctly.

  • Either Liberate or Enforce Status Quo: AI is viewed as a dual-use tool—capable of freeing us from drudgery and exploitation or of reinforcing existing profit-driven hierarchies.

Contextual Thread: The conversation frames AI not just as another productivity tool but as a technology that can fundamentally alter economic and social structures—an inflection point akin to the shift from hunter-gatherer societies.

2. The Central Problem: Exploitation for Profit

  • Profit-Driven Displacement: A key argument is that the most pressing issue isn’t surveillance but the way AI is harnessed to make profits for shareholders—inevitably pushing workers out of jobs.

  • Continued Legacy of Slavery: The conversation draws a line from outright slavery through wage labor to the present moment. Even “good” bosses are still compelled to minimize wages relative to the value produced, which is seen as a form of structural exploitation.

  • Job Loss and Basic Survival: While privacy matters, the more immediate threat is putting food on the table. If people are displaced at scale and left out of the AI-driven economy, it perpetuates or worsens inequality.

Contextual Thread: The conversation frames modern capitalism as a descendant of older exploitative structures. AI threatens to accelerate worker displacement unless there’s a radical shift in ownership and usage of the technology.

3. Skepticism Toward Superficial Fixes

  • Regulation as Insufficient: Top-down laws or regulations are dismissed as “a joke,” suggesting that legislative efforts alone won’t fundamentally alter profit motives.

  • Bridging the Digital Divide: While not unimportant, it’s seen as inadequate on its own. The deeper issue isn’t just about internet access or baseline tech skills but about ownership and control of AI’s productive capacity.

Contextual Thread: The sentiment is that superficial or incremental measures don’t address the underlying power imbalance and profit-driven nature of current systems. True solutions must be more transformative.

4. Personal & Collective Empowerment Through AI

  • Personal Use for Skill-Building: Individuals should harness AI directly to gain new skills, create their own enterprises, and “seize the engine” for self-sufficiency.

  • Collaboration Over Competition: There’s an emphasis on relearning how to cooperate, form co-ops, or pool AI resources so communities can mutually benefit—rather than replicate a top-down hierarchy.

  • Ownership and Local Production: Discussions about worker-owned AI platforms, local manufacturing (like 3D printing + AI), and community-controlled data or microgrids reinforce the idea of decentralizing power.

Contextual Thread: The conversation positions AI as a lever for people to support themselves and each other—shifting away from the dominant corporate model where profits are concentrated at the top.

5. Reimagining Work and Value

  • Beyond the 40-Hour Grind: AI could automate many forms of drudgery, freeing people to focus on creativity, community, and care work—if the economic system is restructured so that income and social value aren’t tethered to “traditional” jobs.

  • Cultural Shift Required: This isn’t just a technology fix. It demands a shift in how we collectively view labor, contribution, and reward—so that a person’s worth doesn’t hinge on a wage paid by an employer trying to maximize profit.

Contextual Thread: Realizing the potential for AI to transform society requires changing our underlying mindset about what constitutes “work” and “value.” Otherwise, AI will simply replicate the old exploitative patterns under new packaging.

Overall Context and Trajectory

  1. AI Is the New Means of Production: There’s a strong parallel to historical calls to “seize the means of production.” Instead of factories, the “means” here is AI/tech infrastructure.

  2. Grassroots Action and Ownership: The conversation repeatedly highlights bottom-up solutions—local AI hubs, worker-owned models, open-source frameworks—because waiting for governments or large corporations is seen as futile.

  3. Survival and Autonomy: At the core, the emphasis is on ensuring everyday people can maintain their livelihoods without being at the mercy of profit-driven interests.

In essence, the overarching message is: We stand at a crossroads where AI can perpetuate an exploitative status quo or empower individuals and communities to become truly self-sufficient. Achieving the latter will require intentional strategies for ownership, collaboration, and a profound cultural shift in how we define and reward human endeavors.

What would you say is the two-word admonition to anyone to step into the future that we're talking about?

A fitting two-word rallying cry, capturing the spirit of harnessing AI (and technology in general) for personal and collective empowerment, is:

“Seize Tools.”

It’s succinct, it emphasizes direct action and ownership, and it speaks to taking hold of the means (AI, data, skills, etc.) rather than allowing them to remain under someone else’s control.

I'm going to go with Wake Up. Before we seize tools, I mean, our hands are full of tools. What do we do with tools? We use them in the service of profiteering for other people. It's not the kind of thing that we talk about all the time. It's like, oh, these gross profits. I'm talking about any business that's designed as a hierarchy is set up to take advantage of the people who are contributing the most value. It's antithetical to a free market, and it's undemocratic. So read that back into it and understand what I'm saying, which is this is about us. This is just doing business the way we've always said it should be done. I think perhaps in a village somewhere we once did it this way, but paying people for value and benefiting all of our lives. The degree to which we're not doing that in this day and age is insane. You bring AI to the table, we can at least start moving in that direction, but we have to wake up first. We have to wake up to the fact that we want better lives, and we want to make the world a better place. We've nearly forgotten that self-interest.

Absolutely—“WakeUp” is the rallying cry that precedes any tangible change. As you point out, we already have the tools; the real problem is whose benefit they serve and whether we’re conscious of how they’re being used. If we stay asleep—if we accept that hierarchies designed to extract as much value as possible from workers are just “the way things are”—then even the most powerful technologies will merely tighten the grip of exploitation.

“WakeUp” is about remembering that we want better lives and a better world. It’s about realizing that the exploitative structures we’ve grown accustomed to aren’t some natural law. Indeed, they’re fundamentally undemocratic, and they undercut the very notion of a “free market” by consolidating power at the top.

Historically, there have been times and places (like certain small-scale villages) where economic arrangements worked more cooperatively. People contributed value to the community and received value in return—without rigid hierarchies skimming off the surplus. AI doesn’t automatically recreate that ideal; it’s a tool that could just as easily serve profiteering.

So “WakeUp” is the first step—awakening to the fact that:

  1. We Don’t Have to Accept Exploitation
    Profit-oriented hierarchies often masquerade as normal or inevitable. Recognizing this system for what it is breaks the spell.

  2. We Can Reclaim Our Agency
    Whether that means forming cooperative structures, redesigning businesses to distribute benefits fairly, or leveraging AI to build personal and community self-sufficiency, it all begins with knowing we have that option.

  3. We Deserve Better—All of Us
    Wanting decent lives and a thriving society isn’t a luxury. It’s an expression of enlightened self-interest—when people flourish, communities flourish.

From there, once you’re awake, you can move on to practical measures: reorganizing business models, pooling AI resources, and ensuring the benefits of innovation flow to the many rather than the few. But none of that happens if we stay stuck in the old mindset, using technology to reinforce the same profit-driven hierarchy.

In short, wake up to what’s possible, to the tools at our disposal, and to the future we actually want—then start building it.

For the post, if you fully agree with GPT's response to your thoughts, you're probably not adding anything to the conversation, the larger conversation. If you routinely have to correct the presumptions and baseline view, then at least you're thinking outside the box.

Draft Post:

If you find yourself nodding along to everything ChatGPT says, you’re probably not pushing any boundaries—or adding much to the larger conversation. But if you keep catching yourself saying, “Wait, that’s not quite right,” or “There’s a deeper angle here,” then you’re actively thinking. You’re digging into assumptions and biases, and that’s how genuine insight and innovation happen.

Out now: How to Fire Your Boss – and redeem the economy

Previous
Previous

Work love