Google's chatbot, Bard, was announced in February and aims to compete with OpenAI's ChatGPT, which Microsoft wants to integrate into its Bing and Edge products.

Google gives us the first glimpses of their new AI


A moment awaited by many, Google has just lifted the veil on Bard, a revolutionary innovation in the field of AI. Bard is an experimental LaMDA AI-based conversational service designed to rival ChatGPT. This new chatbot seems to be up to all the demands and Google is already giving us an exciting preview.
When we ask the question "what discoveries from the James Webb Telescope can I tell my 9-year-old son" Google's chatbot knows exactly where to take you. With a precise and adapted answer, this chatbot answers your question in a simple and clear way. The answer is structured in three parts, with a concise, yet effective explanation.

Google Bard official: When did it happen?

On February 6, Google officially announced the existence of Bard, which was followed by a conference on artificial intelligence organized by the search giant in Paris on February 8. This week has been full of AI-related events, as in addition to Bard's announcement and the conference in Paris, Microsoft also announced its own AI developments with ChatGPT and OpenAI on February 7.
The press had already echoed the existence of Bard at the end of January, when media had been able to observe its functioning as a chatbot called "Apprentice Bard". Differences with ChatGPT have been noted, particularly with regard to the relevance of the information provided.

But what are the differences between Bard and ChatGPT?

Google presents an innovative technology with Bard, which differs from ChatGPT in terms of the relevance of the answers. Unlike ChatGPT which is based on outdated 2021 data, Bard is connected to the web in real time, allowing it to process newer information to respond to user requests in an accurate and timely manner.
While ChatGPT is based on OpenAI templates and GPT-3 technology, Bard is powered by another language model developed by Google: LaMDA (Language Model for Dialogue Applications). This shows a stark difference in how the two technologies are built to help users get information and answers.Google presents an innovative technology with Bard, which differs from ChatGPT in terms of the relevance of the answers. Unlike ChatGPT which is based on outdated 2021 data, Bard is connected to the web in real time, allowing it to process newer information to respond to user requests in an accurate and timely manner.
While ChatGPT is based on OpenAI templates and GPT-3 technology, Bard is powered by another language model developed by Google: LaMDA (Language Model for Dialogue Applications). This shows a stark difference in how the two technologies are built to help users get information and answers.

How does Google Bard work?

oogle Bard is based on the LaMDA language model, which stands for Language Model for Dialogue Applications. First introduced to the general public in 2021, LaMDA is more capable thanks to an update. It is based on an artificial neural network called Transformer, developed in 2017, as well as research in artificial intelligence dating back to 2020.
Google describes Bard as an "experimental conversational AI service," and while Bard isn't yet publicly available, it brings a practical application to work on the LaMDA language model and the Transformer artificial neural network. Bard is also connected to the Internet to have access to always up-to-date information.
The American company announced that the current version of Bard uses a lighter version of LaMDA. “This model, being smaller and requiring less computing power, allows us to deploy to a larger number of users and process a larger volume of feedback,” Google said during the presentation of its chatbot.

Are you a professional and looking for a communication agency for your project? Make an appointment with our team to discuss your project and discover our tailor-made working method.