The Attention Crisis: Language, Meaning, and the Architecture of Augmented Human Intelligence
When infinite language collides with finite attention, shared reality fragments. This paper proposes Augmented Human Intelligence as the architectural solution to preserve human judgment in an age of machine-generated content.
For the first time in history, humanity produces more language than we can process: 15–70 trillion tokens daily. This paper argues we’re entering a global attention crisis where infinite language overwhelms finite human attention, destabilizing the intersubjective reality that enables cooperation, democracy, and shared meaning.
Core Insight
The solution isn’t Artificial General Intelligence as an oracle, but Augmented Human Intelligence—systems designed to strengthen human judgment rather than replace it. AHI treats attention as the scarcest cognitive resource, helping us allocate it wisely and preserve our capacity for reflection.
Abstract
Humanity’s most significant evolutionary advantage has always been language: the ability to create shared meaning, coordinate at scale, imagine futures that do not yet exist, and cooperate in ways no other species can. But for the first time in history, the rate at which language is produced has exceeded the rate at which humans can meaningfully process it. We now generate an estimated 15–70 trillion tokens of text per day. Large language models accelerate this further by producing new content at negligible cost and at speeds that dwarf anything humans can match. The result is not simply information overload. It is the erosion of our intersubjective reality: the shared layer of beliefs, norms, meaning, and trust upon which all societies depend.
This paper argues that we are entering a global attention crisis: a mismatch between the infinite production of language and the finite capacity of human attention. This mismatch destabilizes the foundation of shared knowledge that democratic institutions require, corrodes trust, accelerates the spread of manufactured realities, and overwhelms the neurological systems that support deliberate thought. Historical precedents, from the Malleus Maleficarum to the Rohingya genocide, reveal a recurrent pattern. When new language technologies outpace societal adaptation, the resulting distortion of shared reality enables harm at scale.
Yet there is a path forward. Instead of pursuing Artificial General Intelligence as an oracle of truth, we argue for Augmented Human Intelligence (AHI): systems designed not to replace human judgment but to strengthen it. AHI treats attention as the scarcest and most valuable cognitive resource. It provides context, identifies manipulation, surfaces what matters, and widens rather than narrows our informational horizons. In doing so, AHI supports the most fragile yet crucial component of human cognition: the prefrontal cortex’s capacity for reflection, restraint, and wise action.
How This Fits
Applies attention theory to our current societal crisis, showing how modular AI architectures (Beyond Scale) can be designed specifically to augment human cognition rather than overwhelm it, creating the foundation for practical AHI systems.