Google launches its own ChatGPT: it’s called Bard and it’s based on LaMDA, its controversial AI


After weeks of speculation, Google it has finally been encouraged to get into the battle of artificial intelligence chatbots. The Mountain View firm has just announced Bardyour answer to ChatGPT based on LaMDA, its powerful—and controversial—natural language model.

The announcement of this new tool was made by Sundar Pichai, the CEO of Google, through a publication on the company blog. More details about its capabilities are expected at an event on the integration of artificial intelligence in the search engine that the company will hold this Wednesday.

According to the executive, Bard is available from today to a small group of “trusted testers”. However, the alternative to ChatGPT will be freely accessible to a wider audience in the coming weeks.

From the little that has been seen in the shared images, the operation of the Google chatbot will be very similar to that used by ChatGPT. Users will be able to ask all kinds of questions, and the platform will take advantage of LaMDA to provide quick and easily understandable answers.

Pichai explained that, in principle, Bard will be launched using a lightweight version of LaMDA. This will allow the platform to function without requiring the full computational power that powers its natural language model. The executive indicated that they will take advantage of public feedback and their own investigations to offer “a high level of quality, security and robustness in real-world information.”


Bard is Google’s alternative to compete with ChatGPT

One of the aspects that Google highlighted about its ChatGPT competitor is the ability to get information from the web. Something that places it in an interesting position, considering that the OpenAI chatbot does not extract information in real time from the Internet, but rather its knowledge is limited to events until 2021.

Bard seeks to combine the breadth of the world’s knowledge with the power, intelligence, and creativity of our great language models. He draws on information from the web to provide fresh, high-quality answers,” explained Sundar Pichai.

But Google doesn’t want users to just interact with Bard in isolation.. Those from Mountain View promise to integrate it into their popular search engine, to get even more out of the capabilities of their powerful language model. Basically, Google Search will do with Bard what Microsoft intends to do with ChatGPT on Bing.

“Soon, you’ll see AI-powered features in the search engine that pull complex information and multiple perspectives into easy-to-digest formats, so you can quickly understand the big picture and learn more from the web—whether it’s looking for additional perspectives, like blogs from people who they play the piano and guitar, or delving into a related subject, as steps to get started as a beginner. These new AI features will begin rolling out to Google search soon.”

Sundar Pichai, on the integration of Bard in the Google search engine.

Google’s change of position

Bard |  ChatGPT |  Google

Bard’s announcement is not surprising, but it does mark a dramatic shift in position at Google. In the middle of last December, those of Mountain View they assured that they would not go out of control to launch a competitor of ChatGPT. His argument was that they were exposing themselves to a great “reputational risk”, especially if their chatbot started delivering answers that were incorrect, or outright false. Although the ghost of the controversy over LaMDA and the engineer who assured that the AI ​​had become conscious also flew over.

However, it is clear that the landscape has changed enough in the last 45 days for the company to change its mind. Microsoft’s multi-million dollar investment in OpenAI, added to the integration of ChatGPT and other technologies of the startup in their products, possibly had a lot to do with this story.

It will also be interesting to see how developers can interact with Bard to implement it in their own apps or tools. Sundar Pichai indicated that access to its generative language API will be provided soon, initially potentiated by LaMDA. But he clarified: “Over time, our goal is to create a set of tools and APIs that make it easier for others to build more innovative applications with AI.”

subscribe to Hyper*

Receive our newsletter every morning in your email. A guide to understand and obtain the keys to what is really important in relation to technology, science and digital culture.

Processing…

You did! You are already on the list.

Oh! An error has occurred and we were unable to process your subscription. Please refresh the page and try again.


Related News

Marvel’s Spider-Man: PC remastered features revealed

Sony has announced the features that will be implemented with the Remastered for PC of Marvel's Spider-Man, title developed by Insomniac Games and originally

OnePlus 10T confirmed for August 3 with flagship chip

Pete Lau, CEO of one plusannounced with a post on his company's forums that the OnePlus 10T It will be officially announced on August 3 at an event to be held

ZTE announces nubia Z40S Pro with Snapdragon 8 Plus Gen 1

ZTE announced in China nubia Z40S Proan update to the nubia Z40 Pro launched in the first half of 2022 with a Snapdragon 8 Gen 1 chip. The ZTE nubia Z40S Pro

Carbon, a new programming language from Google, aims to be the successor to C++

Carbon, the latest programming language to be produced within Google, was unveiled today as an experimental successor to C++.

How to activate the renewed function of temporary messages in WhatsApp

WhatsApp it continues to have a series of updates that benefit not only its most recent users, but the entire world. Through the application you can react

WhatsApp: what does the emoji of the yellow face with a monocle mean

Does it look familiar to you? WhatsApp It is one of the favorite platforms for more than 12 billion people in the world. Through it you can send unlimited