Google disputes the claim that it trained its AI chatbot Bard using ChatGPT's data

Last month, in a very disorganized manner, Google announced Bard, its own AI Chatbot. Soon after it was released Bard a chatbot designed to compete with ChatGPT the popular one, received criticism for its inaccuracy.

In its opening advertising, Bard had a factual error and Reuters had also spotted it. Then, a group had left Google embarrassed during a press conference in Paris when a demo phone disappeared. The introduction of ChatGPT, according to earlier accounts, may have helped to create a "Code Red" scenario at Google's headquarters.

Google denies training Bard on ChatGPT data

The Information revealed recently that Google trained Bard using the data from ChatGPT. But, the tech giant denied these claims in a response to The Verge. According to spokesperson Chris Pappas, "Bard is not trained on any data from ShareGPT or ChatGPT. It was claimed that ShareGPT a website provided Google with data from ChatGPT.

Jacob Devlin a former Google AI engineer, left the business to work for OpenAI, as was also reported in the Information piece. According to reports, the former Google employee "warned Google to not utilize ChatGPT's data as it would violate OpenAI's service restrictions." Also, a source informed The Information that following the former employee's 'warnings,' Google ceased utilizing ChatGPT's data.

Bard rolled out for public testing

Google just released Bard for public testing in the meanwhile. Before, the general public couldn't utilize the chatbot. Google, however, just stated that Bard will be made accessible for testing to a small number of users in the US and UK. The latest AI chatbot in town requires users to sign up for the waitlist and wait to gain access.

In its blog post announcing the public release of Bard, Google stated: "Bard may help you be more productive, develop ideas more quickly, and fulfill your curiosity. You may ask Bard to provide advice on how to achieve your goal to read more books this year, to explain quantum physics in plain language, or to inspire your creativity by providing an outline for a blog post. Testing Bard has taught us a lot thus far and getting input from additional users is the next important step in making it better."

Bard is brand-new to the field of Artificial Intelligence and is Still Undergoing Development, so it will sometimes make mistakes. In an email to staff members, Google CEO Sundar Pichai also forewarned them of Bard's potential errors, according to a CNBC article.

"They'll wow us when more people begin to utilize Bard and test its possibilities. There will be problems, "In the Email to Google staff, Pichai said.

Employees who helped test Bard

In the same email, CEO Pichai also said that 80,000 Google staff members assisted with internal Bard testing. The message stated, "I'm grateful to the Bard team, who has spent the previous two weeks working with Bard more than anybody else. And really thankful to the 80,000 Google employees who part in the company-wide Dog Food Testing. We should be proud of this effort and the years of Technological Advancements that got us here, including our research on the 2017 Transformer and key models like PalM and BERT."

Here is what we know about Microsoft's AI Chatbot Bing Getting Advertisements Shortly. In an effort to compete with ChatGPT, Google AI Bard makes a mistake in its first presentation. Google's stock drops Elon Musk says that users of Twitter who do not pay for Blue Tick would lose access to two functions.