Concept: Language model

Generating the language of AI harms: mapping guardrails using critical code studies

Analyzing how major AI companies embed safety rules into their language models