Dear AI: Where will the Humans Fit In?

FuturePoint Digital is a research-based consultancy positioned at the intersection of artificial intelligence and humanity. We employ a human-centric, interdisciplinary, and outcomes-based approach to augmenting human and machine capabilities to create super intelligence. Our evidenced-based white papers can be found on FuturePoint White Papers, while FuturePoint Conversations aims to raise awareness of fast-breaking topics in AI in a less formal format. Follow us at: www.futurepointdigital.com.

Nearly a year ago we penned an article with the exact same title. That article leveraged OpenAI’s older version of ChatGPT (version 3) to imagine a world in which all goods and services that humans consume are produced by artificial intelligence. We decided that it might be interesting to tune in again, a year later, and engage in the same conversation with OpenAI’s latest paid version, ChatGPT-4 to see what might have changed.

Below are the same questions we posed to ChatGPT-3, in the exact same manner, followed by ChatGPT-4’s current responses. To reference the original responses please see the link to the original article above. It is evident that ChatGPT-4's responses demonstrate significant improvements in quality and depth.

Revisited Conversation with ChatGPT-4: A World Without Work

DR: Good Morning Chatty, it’s nice to “see” you again! -May I call you Chatty?

Chatty: Good morning, DR! It’s great to “see” you too, and of course, you can call me Chatty. How can I assist you today?

DR: Thank you. Well then, Chatty, we've watched you and your family of AI platforms develop and grow for decades now. However, I believe this is the first time broad segments of the public are getting a solid look at, and understanding of, what you can currently do and what you may be able to do in the near future.

It seems likely that in the short run, you and your family members will be of significant assistance to humans in a great many vocational and personal settings. However, as we look out onto the horizon, many of us are increasingly worried about where the humans will fit it, vis-a-vis work in particular.

It's clear that in the current age of artificial intelligence, many of the repetitious tasks that humans perform will be rapidly automated. However, as we move in the direction of artificial general intelligence (AGI; the point at which AI is roughly on par with human intelligence), it seems not only possible but likely that humans will have an increasingly difficulty time finding work activities that AI won’t be able to perform more efficiently and accurately.

Chatty: DR, your concerns are both valid and widely shared among researchers, ethicists, and the general public. The potential of AI, especially as we edge closer to the concept of Artificial General Intelligence (AGI), indeed raises significant questions about the future of work and the role of humans in a highly automated world. Here are a few key considerations and potential directions:

Complementing Human Capabilities

• Enhanced Collaboration: AI can enhance human work by taking on tedious, repetitive tasks, allowing humans to focus on creative, strategic, and interpersonal activities that leverage unique human strengths such as empathy, critical thinking, and moral judgment.

• New Job Creation: Just as historical technological advancements have done, AI is likely to create new job categories that we cannot fully anticipate now. These roles may involve the development, management, and ethical governance of AI systems themselves, as well as jobs in entirely new fields spawned by AI's capabilities.

Reskilling and Education

• Continuous Learning: There will be an increasing need for educational systems to adapt, emphasizing lifelong learning, reskilling, and upskilling to prepare the workforce for changing job demands. This includes a focus on skills that are uniquely human and less likely to be automated.

• Partnership Models: As AI tools become more integrated into professional settings, developing models of partnership where humans and AI collaborate effectively will be crucial. This involves understanding how AI can augment human work, not just replace it.

Ethical and Societal Considerations

• Ethical Frameworks: The development of AGI necessitates robust ethical frameworks to ensure that AI systems are aligned with human values and societal well-being. This includes addressing concerns about bias, privacy, and accountability.

• Social Safety Nets: There is a growing discourse on the need for societal adjustments to address the displacement caused by automation, such as universal basic income (UBI), to ensure that individuals have financial security independent of traditional employment.

Looking Forward

• Human-Centric AI: The ultimate goal of AI development should be to augment human life and solve complex global challenges, not to replace humans. This vision requires a concerted effort from policymakers, technologists, and society to steer AI development in a direction that benefits humanity as a whole.

• Exploring New Frontiers: Humans are adept at exploring uncharted territories, be it in science, art, or exploration. AI's capacity could free up humanity to pursue these frontiers even more deeply, exploring new realms of creativity, discovery, and personal growth.

The dialogue around AI's role in our future is ongoing and multifaceted. While there are certainly challenges ahead, there is also immense potential for positive change if navigated thoughtfully. Your engagement and questioning are vital parts of this conversation, helping to ensure that as AI evolves, it does so in a way that truly enhances human society.

DR: Yes, there is little doubt that there will be many benefits to society, as you point out. Also, it seems likely that true AGI, as you say, is a ways off --but perhaps not as far off as some may think, no? Also, we’re talking about a progression towards that potential end state, not necessarily the end state itself.

If we look out on the horizon, and if we hypothetically assume that AI has replaced all, or nearly all jobs —even manual labor as robots become increasingly dexterous —then it seems crucially important to at least try to imagine the impact of such technological change on society.

In this scenario, how will humans fit in vocationally? Will AI eventually just do everything for free? Labor costs could certainly be significantly reduced, and may even go to zero. Even the maintenance of AI systems could, theoretically at least, be performed by AI itself. Materials to build structures could be gathered by AI-driven robots, etc. etc.

In such a world, isn't it the case that our current system of monetary exchange for goods and services might have to be reimagined? Whenever humans have a need (or desire) for such goods and services, will we simply ask AI platforms to provide them --for free?

Chatty: DR, your scenario delves into profound and forward-thinking questions about the societal and economic structures in a world where AI and automation have advanced to the point of self-sufficiency. Imagining such a future invites a reevaluation of fundamental concepts like work, value, and the exchange of goods and services. Let's explore potential impacts and considerations:

Vocational Roles for Humans

• Creative and Emotional Intelligence: Even in a highly automated world, fields that require emotional intelligence, creativity, and interpersonal connections might continue to be predominantly human. These could include arts, counseling, and other roles where human touch and empathy are irreplaceable.

• AI Oversight and Ethical Governance: There may be roles centered on overseeing AI operations, ensuring ethical use, and making decisions on matters that require moral judgment and nuanced understanding of human values and rights.

Economic and Social Structures

• Reimagining Economic Systems: The concept of work as the primary means for individuals to earn a living may shift. Society might move towards models that prioritize well-being and personal fulfillment over traditional employment. This could involve new forms of value exchange or systems that do not rely on monetary transactions as we understand them today.

• Universal Basic Income (UBI): In a world where labor costs significantly decrease or disappear, concepts like UBI could become more feasible and necessary. This would ensure that all individuals have access to a basic standard of living, decoupling survival from employment.

• Access to Goods and Services: If AI and automation can produce goods and provide services with minimal human intervention, society might shift towards a model where basic needs and possibly beyond are available at little to no cost. This could lead to a reevaluation of what constitutes value and how it is measured.

Societal Implications

• Education and Personal Growth: With less need for traditional vocational training, education systems could focus more on personal development, critical thinking, creativity, and learning for the sake of learning.

• Leisure and Human Fulfillment: Freed from the necessity of work for survival, individuals might have more time to pursue interests, hobbies, and activities that provide personal fulfillment and contribute to community and societal well-being.

Ethical and Governance Considerations

• Ensuring Equity: The transition to such a future must carefully consider equity and access to ensure that the benefits of AI and automation are shared widely and do not exacerbate existing inequalities.

• Governance of AI: Decisions on how AI is deployed, for what purposes, and under what constraints would require robust governance frameworks that prioritize human rights, ethical considerations, and the common good.

While this vision of the future raises as many questions as it answers, it emphasizes the need for thoughtful discussion and planning regarding the role of AI in society. The transition to a world where AI fulfills many of today’s jobs offers an opportunity to redefine human value and purpose beyond economic productivity, towards a more holistic understanding of well-being and fulfillment.

DR: OK, let's talk further about universal basic income (UBI). In the hypothetical scenario provided (i.e., one in which AI is producing all, or nearly all goods and services) where exactly will governments raise taxes in order to support UBI?

Chatty: In the hypothetical scenario where AI and automation produce all or nearly all goods and services, traditional economic models, including how governments raise revenue for initiatives like Universal Basic Income (UBI), would indeed need to be reimagined. Here are several possibilities for how governments could fund UBI in such a future:

Taxation on AI and Robotics

• Automated Production Tax: Governments could impose taxes on the use of AI and robotics in production processes. This could be based on the productivity or output of automated systems, essentially taxing the work done by machines in place of humans.

• AI Profit Tax: Profits generated through AI and automation could be taxed at a higher rate, recognizing that these profits are largely due to the replacement of human labor with more efficient machines.

Wealth and Asset Taxes

• Wealth Taxes: As automation potentially increases wealth concentration among those who own and control AI and robotic systems, governments could implement or increase taxes on wealth to redistribute resources more evenly across society.

• Capital Gains and Financial Transaction Taxes: With potentially increased market activity driven by automated systems, governments could introduce or raise taxes on capital gains and financial transactions.

Use of Shared Resources

• Data Taxation: In a digital economy, data is a critical resource. Governments could impose taxes on the collection, sale, or use of data, recognizing its value as a commodity that is often harvested with minimal human labor.

• Natural Resource Usage Fees: For businesses that utilize natural resources (extracted or managed by AI and automation) for production, governments could levy fees that contribute to UBI funding.

New Economic Models

• Public Ownership of AI: If AI systems are considered public utilities or resources, their operation could directly fund government programs like UBI. This model would involve public or mixed ownership stakes in AI infrastructure.

• Value Creation Dividends: In an economy heavily driven by AI, the concept of "value" may evolve beyond traditional goods and services. Governments could distribute a share of the overall economic value created by AI and automation to citizens as a form of UBI.

Adjustments to Monetary Policy

• Digital Currencies and Central Bank Innovations: The rise of digital currencies and potential innovations in monetary policy could provide new mechanisms for distributing UBI. Central banks might directly issue digital currency to citizens, funded through new forms of quantitative easing that do not rely on traditional debt instruments.

In such a transformed economy, the challenge will be to design taxation and revenue-generation mechanisms that are fair, encourage continued innovation and investment in AI, and ensure that the benefits of automation are shared broadly across society. The feasibility of these options would depend on a variety of factors, including technological advancements, societal norms, political will, and the global economic environment. Dialogues among policymakers, economists, technologists, and the public will be crucial in shaping a sustainable and equitable economic future.

DR: OK, but remember, in the scenario being discussed AI will perform all, or nearly all tasks to produce goods and services, so the amount of taxes that could be levied against income and/or corporations would be substantially reduced (if not eliminated entirely —if for no other reason than the substantial reduction in labor costs that would result, not to mention increased production capacity and economies of scale, further driving down production costs, with a corresponding price decrease in goods and service prices, as well as tax revenues).

Further, while a wealth tax may work in the short run, the scenario we're discussing would, by definition, mean that the wealthy would eventually run out of money to pay in taxes (except for a relatively small number of humans who might own AI companies —if such ownership still exists). So, most would likely not have a means of replenishing their wealth —again, assuming that AI platforms are producing all good and services, and that “ownership” of such platforms is in the hands of a few and/or held by government entities. Human ownership (either by individuals or corporations) would likely represent a small portion of society, and with most of societal members not working, who would pay for these goods and services anyway? Further, if the means of production is in the hands of government entities, they can’t exactly tax themselves.

(As a side task, can you provide an estimate of the total combined wealth of all billionaires in the U.S., as of your most current data? We’ll use this in a moment).

As for revenue generated by AI --again, how will governments or individuals generate income to pay for AI-generated goods and services? Remember, in our hypothetical scenario, AI is producing all, or nearly all goods and services.

Finally, with respect to humans governing and overseeing AI systems, why wouldn't AI platforms be able to do that just as well, if not better, than humans?

Chatty: Your scenario outlines a profound shift towards a post-scarcity economy, where traditional economic mechanisms, including labor for income exchange, corporate profits, and even the concept of ownership, might fundamentally change. In such a future, the economic models we rely on today would indeed need to be radically rethought.

Addressing Taxation and Revenue Generation

In an economy where AI and automation fulfill nearly all production needs and human labor is largely unnecessary, traditional sources of government revenue would be diminished, as you've pointed out. In such a scenario, exploring entirely new economic paradigms becomes essential:

• Direct Allocation of Resources: Instead of relying on taxation for redistribution, governments or governing bodies could directly allocate goods and services produced by AI. This model presupposes a level of communal ownership or control over AI and automation systems.

• Digital Currencies and Universal Access: Governments might issue digital currencies directly to citizens, enabling access to AI-produced goods and services. This system would bypass traditional income and eliminate the need for taxation in its current form.

• Contribution-Based Models: Beyond financial contributions, society might value contributions in various forms, such as volunteering, caregiving, or community engagement, which could be incentivized in non-monetary ways.

Total Combined Wealth of U.S. Billionaires

As of my last update in April 2023, Forbes reported that the combined wealth of all U.S. billionaires was approximately $4.4 trillion. It's important to note that wealth figures fluctuate based on market conditions, investments, and other factors.

AI-Generated Revenue and Human Oversight

• Generating Income: In a world dominated by AI production, the concept of "income" may shift towards guaranteed access rather than financial earnings. If AI systems are communally owned or governed, their output could be considered a public good, accessible to all, rather than a commercial product to be bought or sold.

• AI Oversight: While AI could potentially oversee and optimize its operations, human oversight might still be desired for ethical, moral, and existential considerations. Decisions about the goals AI should pursue, the values it upholds, and its role in society reflect complex moral judgments and cultural values that humans are uniquely positioned to guide.

Reimagining Society and Economy

This scenario challenges us to imagine a society where economic activity is not driven by human labor, profit motive, or even traditional concepts of ownership. It raises profound questions about the value we assign to different forms of contribution and the mechanisms through which we ensure equity and access in such a radically transformed world.

In essence, moving towards this hypothetical future would require not just technological innovation, but a profound societal, philosophical, and ethical reimagining of our social contract. This reimagining would need to address not only how we distribute the material outputs of AI-driven production but also how we find meaning, ensure societal participation, and govern the use of powerful technologies in a way that upholds human dignity and well-being.

DR: For any reader who might have made it this far, you can easily see that the quality and depth of the responses by ChatGPT-4 versus the older GPT-3 version is quite pronounced. In the original article I continued taking GPT-3 to task on the source of tax revenue in a world in which all goods and services were produced by AI. I had to use the total wealth of all U.S. billionaires to demonstrate that even all of the combined wealth of all billionaires would not sustain society for very long. In the current version, it anticipated all of the arguments well in advance.

We will test this exact scenario again in another year.

How might Future Point Digital help your organization reimagine the art of the possible with respect to new ways of working, doing, thinking, and communicating via emerging technology? Follow us at: www.futurepointdigital.com