From Hindustan Times: 2024-06-10 04:08:20

OpenAI CEO Sam Altman admitted that the company does not fully grasp its product GPT, hindering the release of new versions. He discussed AI safety and its potential benefits for humanity at the ITU AI for Good Global Summit in Geneva. Altman acknowledged that interpreting AI decisions remains a challenge.

Altman highlighted the complexity of AI systems, emphasizing the need to understand their inner workings. Despite this, he stressed that such systems are generally reliable and robust. OpenAI recently introduced GPT-4o and aims to enhance capabilities further with upcoming models on the path to artificial general intelligence (AGI).

Acknowledging the importance of safety and security, OpenAI has established a committee to address escalating concerns. Altman underscored the necessity of developing policies to ensure the safety and reliability of advanced AI models. Understanding the mechanisms behind these systems is crucial for making and verifying safety claims.



Read more at Hindustan Times: Sam Altman says OpenAI doesn’t completely understand how GPT works: ‘We don’t know what’s happening’