will the model let you be offended? And in the past, we've had much tighter guardrails on this. But I think part of model alignment
OpenAI’s Sam Altman Talks ChatGPT, AI Agents and Superintelligence — Live at TED2025
1 example from real videos — listen, replay, loop.
will the model let you be offended? And in the past, we've had much tighter guardrails on this. But I think part of model alignment
OpenAI’s Sam Altman Talks ChatGPT, AI Agents and Superintelligence — Live at TED2025
will the model let you be offended? And in the past, we've had much tighter guardrails on this. But I think part of model alignment
OpenAI’s Sam Altman Talks ChatGPT, AI Agents and Superintelligence — Live at TED2025