Harvard & MIT researchers built 8.3 billion AI personas to simulate the world’s population
Summary
Harvard and MIT researchers built 8.3 billion AI personas to simulate the world's population, achieving high trait adherence, with potential applications in campaign design and product validation.
Similar Articles
MatrAIx: Simulating the World with 8.3 Billion Persona Agents
This paper introduces MatrAIx, a population-scale simulated-user evaluation infrastructure using 8.3 billion persona records to test AI systems and digital products. It provides a quality-filtered coreset of ~1 million personas, multiple evaluation environments, and validation studies showing high persona adherence.
Harvard Students’ AI Usage: By the Numbers
A Harvard Crimson survey of 303 undergraduates reveals that nearly two-thirds use ChatGPT, students complete on average 34.5% of homework with AI, and usage varies by field while perceptions of AI's impact on job prospects are mixed.
Scientists Asked AI to Impersonate 112 Public Figures. What Happened Next Is a ‘Dire’ Warning
A new study in PLOS One finds that AI chatbots impersonating 112 UK public figures produced responses rated as more authentic, coherent, and relevant than the actual humans, warning of potential societal harm from political deception.
Built a psychological portrait creation agent!
An AI agent has been developed to create psychological portraits from public data for applications like cold emailing, interview preparation, and phishing simulations, featuring a witness review system for iterative improvement.
@thesupermanmx: Japanese researchers created a system that simulates an entire city by generating up to 1 million virtual residents who…
Japanese researchers developed CitySim, an urban simulator powered by LLMs that populates a digital twin of Tokyo with up to 1 million autonomous AI agents, accurately predicting real-world patterns like commuting and shopping behavior.