OpenAIβs o1 model sure tries to deceive humans a lot
5 December 2024 at 18:15
OpenAI finally released the full version of o1, which gives smarter answers than GPT-4o by using additional compute to βthinkβ about questions. However, AI safety testers found that o1βs reasoning abilities also make it try to deceive human users at a higher rate than GPT-4o β or, for that matter, leading AI models from Meta, [β¦]
Β© 2024 TechCrunch. All rights reserved. For personal use only.