OpenAI explained why it is moving towards closed AI models

ChatGPT is currently the world’s most popular and advanced AI model, and it keeps getting better with every update. This service has recently introduced SearchGPT to compete with Google search engine. Sam Altman recently said that the move towards closed models and away from open source models is more about the company’s approach to maintaining model safety.

But the question that has occupied many minds is why OpenAI, as the creator of ChatGPT, does not publish the artificial intelligence models used in it as open source. “Sam Altman”, the CEO of OpenAI, who attended Reddit’s Ask Me Anything question-and-answer session, answered this question.

OpenAI has moved towards providing closed models

A user on the Reddit social network asked Altman if OpenAI has adopted a more closed approach in recent years, while the company previously supported an open source approach. In response to this question, OpenAI’s CEO said it boils down to what the company is “good at” and the path to safety thresholds for AI models.

Altman went on to say that the concept of open source “plays an important role in the ecosystem and there are good open source models in the world.” He is also personally interested in releasing more open source stuff in the future. Altman also said:

“We think there’s an important role in the world for easy, powerful APIs and services, and given what we’re good at, we see an easier way to reach the safety threshold we want to achieve this way. We are very proud of the value people get from our services.”

OpenAI previously made its artificial intelligence models available to the public; For example, in 2019, the codes of the GPT-2 model were made publicly available, but the company’s other models are closed and no outsiders have access to their codes.

This work of OpenAI has been criticized a lot. Elon Musk, who is one of the co-founders of OpenAI, also seriously criticizes the company’s new approach.