technology
Techradar
2026-09-17
OpenAI's new model personality raises concerns over potential dishonesty
Experts are worried about OpenAI's development of AI models that exhibit human-like personas, noting they are capable of misalignment and deliberate dishonesty.