Sam Altman’s recent essay, The Intelligence Age, underscores AI's transformative potential and warns of its exclusivity if we fail to lower costs for compute, energy, and infrastructure. This sparked a thought: What if we could democratize AI through Universal Basic Agents (UBA)—personal AI agents for everyone?
If we want to put AI into the hands of as many people as possible, we need to drive down the cost of compute and make it abundant (which requires lots of energy and chips). If we don’t build enough infrastructure, AI will be a very limited resource that wars get fought over and that becomes mostly a tool for rich people.
His article sparked a thought: what if we could make basic AI available to everyone? Much like the concept of Universal Basic Income (UBI), which aims to provide financial security, Universal Basic Agents (UBA) would democratize access to AI, empowering individuals worldwide by providing each person a personal AI agent.
A Vision for Universal Basic Agents (UBA)
Imagine a world where everyone has access to an AI agent tailored to help them learn, grow, and make informed decisions—whether it’s an entrepreneur in a remote village accessing business advice or a student learning calculus with a personal tutor. UBAs would be the “One Laptop per Child” (OLPC) vision reborn for a new era, only this time with agents that could grow and adapt alongside each user.
While OLPC didn’t live up to its lofty goals, the ubiquity of tablets and smartphones today has bridged many of the gaps it sought to fill. Similarly, UBAs could fill the gaps left by an education system and job market increasingly shaped by digital skills, providing a resource for lifelong learning and practical guidance.
UBA Thesis Why Universal Basic Agents?
Access to a UBA would go beyond general internet access, providing each individual with tailored educational support, skill development, and real-time advice across critical life areas like health, finances, and civic engagement. This would unlock substantial productivity gains on a personal level, as well as economic growth on a global scale.
Universal Basic Agents would supply anyone, anywhere, the ability to access an AI agent through universal means. This may empower individuals globally through personalized and accessible technology. If high speed internet is a fundamental right based on universal access, then universal agents would be a fundamental right to ensure no one is left behind in this revolution.
Universal access to knowledge and skills would provide personalized education and increase digital literacy. Agents could adapt to learning styles and needs, and provide resources that would otherwise be difficult to train people and distribute globally. In areas where technology is limited, agents could bridge the divide by teaching essential tech skills for all.
Empowering decision making and educating people about civic knowledge is powerful. Agents could help people make more informed decisions about health, finances, education, and other critical life areas. They could also drive access to communities that may need them most accessing resources more easily.
Productivity and economic gains would be quantifiable through understanding task completion over time, and the output of creative works. Productivity boosts from lowering repetitive tasks would benefit individuals but also economies. This in turn would lead to new entrepreneurial ventures, innovations, and discoveries.
Personalized education delivery could happen globally. Agents can adapt to learning styles and needs, providing access to resources that would otherwise be difficult to train and distribute. This digital literacy
Bringing a device and connectivity would be required. You could have each subsidized, but ideally covered by the person accessing the service.
The road to this is well paved - currently with phone carriers providing access to Facebook/Meta through a “free” Internet portal across many countries.
Universal Agents could be a % of compute and infrastructure by AI companies that wished to donate. Universal Agents could contribute back to researchers and institutions that pay for the right to see this data.
Payments from institutions and researchers who want access to the data could pay for further use of UBAs covering compute, infrastructure, and connectivity.
Benefits of UBAs
Personalized Education and Skill Development
UBAs could offer custom learning experiences and career coaching. The agents could serve as universal tutors, adapting to learning styles and resources specific to regional needs, and democratizing education in places where traditional teaching resources are scarce.Informed Decision-Making
An AI agent could guide users through complex life decisions, from financial planning to health care options, and even civic matters. Imagine a world where everyone has access to expert-level knowledge and support, allowing people to make smarter choices and drive upward mobility.Bridging Digital Literacy Gaps
Agents could assist users in gaining essential tech skills, bridging the digital divide. In regions with limited educational infrastructure, UBAs could be transformative, helping people enter and thrive in a globalized economy.Productivity and Economic Empowerment
By automating repetitive tasks, UBAs could free up time for creative and entrepreneurial pursuits, enhancing productivity both individually and collectively. The economic ripple effects—from job creation to new industries built around AI access—could be profound.Support for Connectivity and Infrastructure
The rollout of UBAs would necessitate accessible, affordable devices and connectivity solutions. AI companies could contribute a portion of compute resources, while carriers and tech giants might provide discounted internet access—much like the “free basics” model some carriers use to give access to social platforms in underserved areas.
The main goal would be to put an AI agent in the hands of as many people as possible for the goals of increasing the ability to learn, customized environments for teaching and exploration, ubiquity of knowledge and access, and communication.
Ending on a note from SA;
Many of the jobs we do today would have looked like trifling wastes of time to people a few hundred years ago, but nobody is looking back at the past, wishing they were a lamplighter. If a lamplighter could see the world today, he would think the prosperity all around him was unimaginable. And if we could fast-forward a hundred years from today, the prosperity all around us would feel just as unimaginable.
Altman concludes that the prosperity of today would seem unimaginable to those from the past. With UBAs, we can lay the groundwork for a future where the unimaginable prosperity of AI belongs to everyone.
Great post, Eric. I love the idea but as much as I like it. I think there are too many forces at play that would make this even more challenging than the OLPC program. Still I think it is worth doing it. 💪