OpenAI hired at least 12 Google employees to work on ChatGPT before the AI tool went viral

OpenAI seems to have hired a lot of Google researchers to work on AI tools and this reportedly happened way before the launch of ChatGPT. 

People who are familiar with the matter told The Platformer that more than 12 employees that worked on AI-related projects for Google were hired by OpenAI.

The cited source claims that at least five former Google researchers worked on ChatGPT project and helped OpenAI ready it for the launch. The service was announced back in November last year. 

The information seems to be true considering the researchers that left Google sometime back have been given credits on OpenAI’s blogpost when the company announced the launch of ChatGPT.

Barret Zoph, Liam Fedus, Luke Metz, Jacob Menick, and Rapha Gontijo Lopes are some of the former Google researchers who have been hired by OpenAI.

Some of the employees decided to left the big tech company because the culture had become lethargic, according to the details revealed by two former Google employees who used to work in Alphabet’s AI division.

It is being said that the researchers at Google didn’t get much of a free hand over new product initiatives and there were reportedly a lot of restrictions or issues, which forced people to look elsewhere.

Well, this didn’t stop Google from launching its own AI tool – Bard – to give a tough competition to OpenAI’s ChatGPT. Though, the service is not available right now.

The creator of ChatGPT has asserted in a podcast with NYT that the tool was not designed to be used and the company clearly was not expecting a huge success or the service would be used by a lot of people because the website has gone down several times because it is not able to handle high-traffic.

While it has been proven that ChatGPT can help finish a lot of your work, it has also been discovered that the tool can help pass the toughest medical licensing exam of the US. 

According to details revealed by researchers, the AI tool can score between 52.4 percent and 75 percent in the exams, which is quite impressive. Though the service is weak in Math.