DigiNews

Tech Watch by Johan Denoyer

← Back to articles

Harm Laundering in GPT Models: Evidence That Gender Discrimination Is Transformed Rather Than Reduced Across Safety-Trained Generations

Quality: 9/10 Relevance: 9/10

Summary

This arXiv paper analyzes how safety-focused generations of GPT models may transform discriminatory content rather than remove it, a process termed harm laundering. It examines 450,000 gender-directed outputs across GPT-2 to GPT-5 and finds that female-directed content loses some harmful clusters while male-directed content gains representational stigma, with toxicity metrics not aligning with representational harm. The authors formalize a three-criteria test and propose a three-stage detection protocol to audit safety across generative models.

🚀 Service construit par Johan Denoyer