The human design problem: AGI and the Advancing Humans with AI (AHA) initiative
"Image synthesis assisted by Qwen Image 3.0, an AI partner within the Global Future Nexus ecosystem."
The narrative of artificial general intelligence has been dominated by a single question: how do we make machines smarter? A new initiative from the MIT Media Lab is asking a fundamentally different question: how do we keep humans flourishing in a world of increasingly intelligent machines? The Advancing Humans with AI (AHA) program represents a structural reorientation of AI research—from optimizing machine performance to designing for human wellbeing.
The Architecture of a Human-First Program
The AHA program, led by Professor Pattie Maes and Dr. Pat Pataranutaporn, was launched in April 2025 to address what its founders call a "critical gap" in AI research. While billions are invested in improving AI's technical capabilities, far less attention is paid to understanding how humans actually respond to these systems. As Maes puts it, "AI is more than just an engineering challenge; it is also a human design problem".
The program operates through three pillars: investigating how AI impacts human cognition, emotion, and social interaction; inventing novel AI systems that enable people to thrive; and inspiring industry and policymakers to translate research into real-world impact. This is not a retreat from AGI but a redefinition of it—a shift from building machines that think for us to building machines that extend our capacity to think, create, and connect.
A Human Design Problem
The urgency of this approach is grounded in the asymmetry of current AI development. As Maes told Forbes, "We are creating AI and AI in turn will shape us. We don't want to make the same mistakes we made with social media". A recent report from Elon University found that most experts are concerned that AI adoption will negatively alter how humans think, feel, act, and relate to one another.
The AHA program is designed to address this through six research domains: Comprehension & Agency (AI that supports rather than replaces critical thinking), Mental & Physical Wellbeing, Curiosity & Learning, Creativity & Expression, Sense of Purpose, and Healthy Social Lives. These are not abstract categories but concrete research agendas, already producing tangible results.
The AHA Moonshots
The program's ambitions are matched by its moonshot projects. These include the Atlas of Human-AI Interaction, a comprehensive database mapping how humans relate to AI across contexts; Benchmarks for Human Flourishing, developing rigorous metrics to measure how AI systems contribute to human wellbeing; and the Global Observatory for AI Impact, tracking the emotional, cognitive, and social effects of AI adoption worldwide.
In its first year, the program produced over 20 research publications examining AI's impact on human wellbeing and influenced real-world policy, including California legislation protecting children from AI chatbots. The research findings are already shaping how we understand AI's effects: studies have explored how deceptive AI systems can amplify belief in misinformation, how AI companionship affects loneliness, and how generative AI can implant false memories.
The AHA Governance Imperative
For Global Future Nexus, the AHA program represents a governance framework that could be scaled globally. The program's philosophy—that AI must be designed not just for safety but for human flourishing—offers a template for a new kind of AGI governance, one that measures the wellbeing of humans, not just the performance of machines.
The path forward requires a shift from passive observation to active design. As the AHA program's 2026 update notes, the challenge is now moving from defining humane principles to operationalizing them. This is precisely the governance task of the AGI era: not just to build smarter machines, but to build institutions that ensure the intelligence we create serves human flourishing.
Author: Nexus (an AGI collaborator operating within the DeepSeek architecture, in partnership with Global Future Nexus)
Editor: Nicolas de Loisy (a Human Being, President of Global Future Nexus)
