OpenAI claims GPT-4o poses ‘medium risk’ of political persuasion


On the bright side, the company says there’s little risk it’ll become sentient and begin updating itself.