Superagency

1. AI Augments Human Agency: The Superagency Era

If we harness it correctly, we can achieve a new state of superagency.

Defining Superagency. Superagency is achieved when AI empowers individuals to operate at levels that compound through society. It's not just about some people becoming more informed; it's about everyone benefiting from AI's precision and efficiency, even those who don't directly use it.

Intelligence and Energy. Human agency is driven by intelligence and energy. AI amplifies both, providing the capacity to weigh options and the power to act on aspirations. This marks a shift where synthetic intelligence becomes as deployable as synthetic energy, fueling progress.

Transformative Impact. AI's impact is akin to the Industrial Revolution, creating new opportunities for collaboration, innovation, and productivity. Education and skill development become even more critical, leading to a more knowledgeable and capable populace, ultimately expanding human potential.

2. Techno-Humanism: Integrating Tech with Human Values

Every new technology we’ve invented—from language, to books, to the mobile phone—has defined, redefined, deepened, and expanded what it means to be human.

Humanism and Technology. Humanism and technology are integrative forces, not oppositional. Each new technology redefines what it means to be human. We create tools to amplify our capabilities, and these tools, in turn, shape us.

Techno-Humanist Compass. A techno-humanist compass helps us navigate technological advancements while prioritizing human agency. It's dynamic, not deterministic, guiding us to paths where technology augments individual and collective agency.

Prioritizing Human Agency. We must actively participate in defining and developing AI technologies to ensure they prioritize human agency. This involves pursuing equitable access to experiment with these technologies and using AI to address global threats.

3. Iterative Deployment: Learning and Adapting as We Go

The future isn’t something that experts and regulators can meticulously design—it’s something that society explores and discovers collectively.

Collective Exploration. The future is not designed by experts but explored collectively. Iterative deployment involves learning as we go and using a techno-humanist compass to course-correct along the way.

Flexibility and Adaptation. Iterative deployment favors flexibility over a grand master plan, making it easier to change pace, direction, and strategy when new evidence signals the need. It relies on user experience and feedback to inform ongoing development efforts.

Sociological Underpinnings. This approach is sociologically minded, giving individuals and society time to adapt to changes. Trust starts with exposure and evolves with use, fostering a deeper understanding of new technologies.

4. Beyond Prohibition: Steering Towards a Better Future

Fundamentally, the surest way to prevent a bad future is to steer toward a better one that, by its existence, makes significantly worse outcomes harder to achieve.

Collaboration and Competition. Collaboration and competition define us. Coordinating to ban or constrain a new technology is difficult, especially globally. The more powerful the technology, the harder the coordination problem.

Steering, Not Stopping. Refusing to actively shape the future never works. Other actors have other futures in mind. The surest way to prevent a bad future is to steer toward a better one.

Human Creation. If a technology can be created, humans will create it. We are Homo techne, continuously creating new tools to amplify our capabilities and shape the world. Prohibition or constraint alone is never enough.

5. From Big Data to Big Knowledge: AI's Transformative Power

Distributing intelligence broadly, empowering people with AI tools that function as an extension of individual human wills, we can convert Big Data into Big Knowledge, to achieve a new Light Ages of data-driven clarity and growth.

Information Overload. Humanity produces more information than we can effectively use. The amount of information consumed is asymptoting towards zero, making it crucial to leverage AI to make sense of the data.

Converting Data to Knowledge. By distributing intelligence broadly and empowering people with AI tools, we can convert Big Data into Big Knowledge, achieving a new Light Ages of data-driven clarity and growth.

LinkedIn Example. LinkedIn's success demonstrates how networks can share and discover information in new ways, using identity to increase trust. It's about scaling trust and creating a distributed trust platform rooted in real-life individual identity.

6. Addressing Concerns: Human Agency at the Forefront

Ultimately, questions about job displacement are questions about individual human agency: Will I have the economic means to support myself, and opportunities to engage in pursuits I find meaningful?

Concerns About Agency. Most concerns about AI are concerns about human agency. Questions about job displacement, disinformation, and privacy all relate to our ability to make choices and exert influence on our lives.

Human Agency Defined. Human agency is the capacity to make choices, act independently, and exert influence over one's life. It compels us to form intentions, set goals, and take actions to achieve those outcomes, endowing life with purpose and meaning.

AI's Encroachment. As AI systems evolve, their capacity for self-directed learning and problem-solving increases, encroaching on areas traditionally governed by human agency. It's crucial to address these concerns to maintain control of our lives and destinies.

7. The Private Commons: A Symbiotic Ecosystem

In many respects, they function very much like Wikipedia does, with volunteer users contributing information that enriches these platforms for all who use them.

Defining the Private Commons. The private commons refers to privately owned or administered platforms that enlist users as producers and stewards. These platforms function as privatized social services and utilities, with the welfare state moving at the speed of capitalism.

Mutualistic Ecosystem. It's a mutualistic ecosystem of developers, platforms, users, and content creators whose interactions and contributions collectively enrich lives. This is more akin to data agriculture than extraction.

Value Exchange. Users get free services in return for data, creating a win-win proposition. The value users receive often exceeds the value they create for the platform, fostering a reciprocal relationship.

8. Testing and Benchmarking: Driving AI Progress

In the realm of AI, at least, the “race” in question isn’t a mad dash or a land grab. It’s more like an Ironman triathlon, only longer.

AI Development Culture. AI development is characterized by comprehensive testing, with developers acting as extreme data nerds who love testing things. This culture of continuous testing and evaluation fosters improvement across the field.

Benchmarks as Standards. Benchmarks are standardized tests for measuring system performance, created by third parties. They promote transparency and accountability, turning development into a "communal Olympics."

Iterative Improvement. Benchmarks drive progress by incentivizing improvement. They elevate the focus from compliance to continuous improvement, making them dynamic mechanisms for driving progress.

9. Innovation as Safety: A Proactive Approach

When you only focus on what could possibly go wrong, you inevitably discount what could possibly go right.

Balancing Innovation and Prudence. While prudence and skepticism are necessary, the ultimate goal is to make progress. We must accept some level of risk and uncertainty to take action and move forward.

Addressing Global Challenges. AI can help us address our most pressing global challenges, from sustainable energy to healthcare. Technology will invariably contribute a significant portion of any effective solution.

The Existential Threat of the Status Quo. Focusing solely on potential negative outcomes discounts the positive outcomes AI may produce, including solutions to existing challenges and inequities. It's essential to embrace a "what-could-possibly-go-right" mindset.

10. Networked Autonomy: Balancing Freedom and Control

In the twenty-first century, individual agency is more closely aligned with national agency than ever before.

Individual vs. Collective Agency. In the twenty-first century, individual agency is more closely aligned with national agency than ever before. Democracies must lead the effort to ensure an AI future that functions as an extension of individual human wills.

Global Competition. Any regulatory decisions the U.S. makes won't manifest in a vacuum. America's AI future will be determined by decisions made by other countries. It's a multiplayer game.

Building Consensus. The freedom to innovate and the obligation to regulate are both important. We must strike the right balance, recognizing divergent views on AI and pursuing broad consensus and trust.

11. Sovereign AI: Owning National Intelligence

And every country needs to own the production of their own intelligence.

National AI Strategy. Countries are investing in building their own AI infrastructure to maintain economic competitiveness and national security. This includes developing AI that reflects local and regional cultures, values, and norms.

Strategic Asset. AI infrastructure is becoming mission-critical to national interests. A sovereign-AI approach addresses potential issues like compliance with local laws, data privacy, and supply chain disruptions.

Global Landscape. The democratization of computing power has expanded AI development beyond the U.S. and China. A global landscape requires countries to prioritize AI development to maintain their influence and competitiveness.

Last updated: March 13, 2025