Mass Intelligence: When Powerful AI Becomes Accessible to Billions

Over a billion people now regularly use AI systems such as ChatGPT and Gemini. This shift marks the emergence of 'Mass Intelligence'—a new era where powerful artificial intelligence, once exclusive to a few, becomes as readily available as a web search. Until recently, access to advanced AI models remained restricted to paying subscribers willing to spend $20 to $200 monthly. The majority of free users encountered older, less capable models with significant limitations. Two obstacles blocked mass adoption: users often did not know which model to select, and the best systems simply cost too much. OpenAI reported that fewer than 7% of paying customers regularly selected their most capable models, even when access was available to them. The landscape is shifting rapidly. OpenAI's GPT-5 functions as an intelligent router, automatically directing simple queries to efficient smaller models while allocating greater computational power to complex problems. Google accelerated this democratization by offering free access to its most advanced systems. The impact was immediate: within days, the share of OpenAI's paying customers using reasoning models jumped from 7% to 24%, while free-tier users with access to the most powerful models grew from near zero to 7%. Economic and technological efficiency now drives this transformation. Processing costs have collapsed dramatically: what once cost $50 per million tokens for GPT-4 now costs just 14 cents for the more capable GPT-5 Nano. Google reports 33-fold improvements in energy efficiency over the past year. These gains make it economically viable for companies to offer powerful AI tools to billions of users without sacrificing profitability. User interfaces have also become far simpler. Complex prompting techniques are no longer necessary; modern models understand plain language instructions and execute sophisticated tasks intuitively. Google's new image editing tool exemplifies this: it processes simple descriptions to modify photographs, all without cost. When technology of this caliber reaches a billion users, the implications will be substantial for every institution.