Last Updated: March 27, 2023, 13:57 IST
Google continues to face the tough questions about its Bard AI chatbot after it was alleged that the company is using private Gmail data to train its chatbot. The concerns came up recently after a researcher using the Bard AI chatbot asked where it was getting training data from, to which the AI chatbot said internal data from Gmail. Now, Google has come out quashing those concerns.
Google has been quoted saying by The Register that it does not train the AI chatbot on text from private Gmail accounts.
The company suggested that like most large language models (LLMs), Bard also has a tendency to give inaccurate or misleading information and the models can do that with full confidence which makes you feel like the information shared is the truth. “This is an example of that. We do not use personal data from your Gmail or other private apps and services to improve Bard,” Google was quoted saying.
The company is clearly trying to give itself a clean chit by laying all the blame on the LLMs here but we must realise that Bard has been developed by Google, so all the concerns that are raised about Bard will obviously be directed at the company.
Google is understandably taking it slow with its AI chatbot, which only made it to select countries last week for the public. Meanwhile, you have Microsoft using OpenAI’s prowess in this arena to integrate the ChatGPT v4.0 into Bing and Edge browser. Google is supposed to be the AI champ in the industry, but everyone can see the company struggling to fight the evolution that ChatGPT has made in quick time.
Microsoft has reportedly seen the web traffic on Bing search go up after it got the ChatGPT v4.0 chatbot working on the search engine. It has garnered hundreds of millions of users, and also has a Plus subscription which gives you more features at a cost.
Read all the Latest Tech News here