SHAH ALAM - Google recently unveiled Bard, their experimental AI chat service a direct rival to OpenAI’s ChatGPT on Monday.
It was reported that before it would be made available to everyone in the next several weeks, Google's AI chatbot will first be evaluated by a small group of users.
Powered by Google's Language Model for Dialogue Applications (LaMDA), Bard will respond with all the data available on the internet.
This is a significant differentiator from ChatGPT, which lacks internet access and can only access data up to the year 2021.
"Bard seeks to combine the breadth of the world's knowledge with the power, intelligence and creativity of our large language models,” Google Chief Executive Officer Sundar Pichai said in a statement.
According to international news outlets, Google claims Bard will be able to distill difficult subjects into digestible, chitchat-worthy chunks.
The goal is to spread knowledge more widely in a manner that is clear and can encourage learning in everyone, even young children.
It is said that Bard can help you explain explain scientific discoveries from NASA's James Webb Space Telescope to a 9-year-old
According to the press announcement, the first iteration of Bard will employ a lightweight model version of LaMDA because it uses less processing resources and can scale to more people. By doing this, Google would be able to gather additional feedback before making the change public.
According to Pichai, Bard will be evaluated within to make sure it satisfies standards for quality, safety, and groundedness.
"In order to make AI safe and practical, Google is committed to treating it properly.”
"This is the strategy Google normally adopts, holding onto powerful AI services until they are confident they are prepared for the general public.” the statement said
Examples include Google's highly effective AI music generator MusicLM and AI image generator Imagen, both of which have not yet been made available to the general public.