Google reveals Bard: a ChatGPT competitor
The timing of Google's “Live from Paris” event in France could not have been more appropriate to announce Bard.
It took place just one day after Microsoft organized an event with Open AI.
Just like GPT chat, Bard can answer user questions.
It can also automate tasks, do word processing, respond to emails, summarize content reading such as an email, a database, and write according to the instructions the user gives it on its interface.
In the future, the Bard chatbot will display results directly on the search engine results page for NORA searches, also known as “unanswered” searches, and will be integrated into Google Assistant.
At first reading, Bard summarizes several sources to give the user informed opinion and answers to their questions without simply displaying the featured snippet from a single source to its Internet users.
What is Bard?
ChatGPT is a AI chatbot used for its ability to provide engaging responses based on existing data, but Google is now going further with Bard.
Google explained that Bard is an experimental conversational AI service, powered by LaMDA (Language Model for Dialogue Applications).
It was announced two years ago, at the Google I/O 2021 conference in France, as the field for its next generation of language and conversation capabilities from the firm Google.
It is now used to create these queries for users with a simple interface.
This advanced AI can access resources on the web to produce “fresh, high-quality” responses, as CEO Sundar Pichai pointed out on Twitter — giving us a list and overview of the latest advancements in conversational AI technology on Google's new interface.
The case of Microsoft's recent acquisition of OpenAI fuels the rumor that the company plans to integrate the technology GPT chat in Bing, while industry giant Google is considering a similar approach.
This could limit any redistribution of web search engine market shares if both companies use these AI advancements.
Source: CNBC
With powerful forces like the giant Microsoft and now Bard investing heavily in research onartificial intelligence, there is great potential to disrupt this already competitive space for the search engine world.
Availability of Bard in France
Google I/O conference
For the moment, countries in the European Union (excluding the United Kingdom), including France, cannot take advantage of Google Bard services. Sissie Hsiao, the vice president of Google, spoke about this subject at the Google I/O conference held on May 10 at the Shoreline Amphitheater in Mountain View, California.
More recently, during a roundtable discussion with journalists, Sundar Pichai, the CEO of Google, gave more information on this issue.
He emphasized that, due to the significant difference in legislation between the various countries, including France, a considerable amount of work is needed to ensure the correct implementation of services. Pichai reiterated the company's commitment to “doing it right.” At the same time, Google France confirmed to Le Parisien that they are working closely with specialists and policy makers to ensure that Google Bard is in line with local standards.
Sign up for the Google newsletter to find out more about future access in each Bard country.
specifications
1. Does the Mountain View firm's AI only provide text responses?
Like the ChatGPT chatbot, Google Bard will initially only respond to text prompts (queries) through its own written productions.
But LamDA isn't the only AI Google has in its drawer, and we can expect to see similar advancements in the world of video and audio soon.
2. Can Bard generate art and images?
Google Bard is a tool that has the potential to create art and images.
Inspired by Google's open source DeepDream project, which some say has a neural network capable of producing bizarre images from photographs, Google Bard builds on this concept with a network that has been trained for 20 years.
With this knowledge, the service can generate works of art from scratch - images and compositions that it has never seen before.
The artist creates a seed or a subject to guide the machine in producing something unique.
Once trained, users can start generating artwork without the need for coding experience or artistic skills. It's a fascinating concept that allows us to explore how we can use technology to allow us to be creators rather than denying ourselves our jobs.
Difference between Bard and ChatGPT
Google announced it shortly after the giant Microsoft announced a $10 billion investment in OpenAI.
The PC manufacturer plans to integrate artificial intelligence into its own search engine, Bing.
So far, Google hasn't said that the Bard chatbot will have AI-style chatbot features to answer questions like ChatGPT, or that it will do more than provide a “textual” summary (response) of search results.
The first potential difference between the two was that ChatGPT was not connected to the internet, which meant that it had very limited knowledge of facts or events after 2021.
Google says that Bard, on the other hand, “relies on information from the web” to provide up-to-date answers.
The difference between Bard and ChatGPT may not last very long: a leak recently showed a version of Microsoft's Bing integrating ChatGPT in English, which suggests that ChatGPT will soon be associated with the Microsoft search engine.
What is the main artificial intelligence on which Google Bard is based?
Two years ago, Google developed its language model, equipped with next-generation linguistic synthesis and conversation capabilities, which served as the underlying technology for the development of Google Bard.
However, the company used a lightweight version of the LamDA language model for this project, which will allow for faster deployment for its dissemination to the general public.
Three principles are at the heart of innovation in the framework of the research project: safety, anchoring and the quality of conversations.
While processing data from a variety of sources on the Internet in real time, Google Bard has also been trained to filter biases that may produce less than ideal results.
It has been trained in English to exceed the computational capabilities of any pre-trained model existing on the market. He has been trained in multiple dimensions and for all system sizes in order to automatically refine the results.
FAQs
Do I need a Gmail email address to use Google Bard?
The Google Bard chatbot allows you to search the web and answer questions without having to constantly go back and forth from one website to another.
What's amazing is that you don't need a Gmail email address to use it; you can simply sign in with your existing email address.
From there, you can access all of Google's services, including search, maps, videos, translations, and more, all of which are on the same page.
Where is Google Bard currently available?
Google Bard is currently available in three languages: English, Japanese, and Korean.
It's accessible in over 180 countries and territories, but it's not available in many places, including the European Union (EU), Canada, Hong Kong, and China.
Bard is not available in the EU due to stricter privacy regulations for internet users:
The EU takes a different approach to artificial intelligence for its users compared to the United States. The EU focuses on managing the potential risks associated with AI, such as bias and discrimination.
Google is investing efforts to develop a version of Bard specially designed for the European market (information only in English)
Sign up for the newsletter on the Google site to find out more.
How do I access Google Bard in France without a VPN?
Google Bard's availability is currently limited to specific countries and territories for Internet users, including Canada and European nations such as the Netherlands, France, Spain, Italy, and Germany, among others.
Despite Google's expansion plans, users currently need a reliable VPN in order to change their IP address and access Google Bard unrestricted from any location.
Sign up for the newsletter on the Google site to find out more, click on this link to this article about the best VPNs available.
Conclusion.
In addition to LamDA, Google says it has other AI concepts, such as PalM, Imagen, and MusicLM, that could create “entirely new ways of understanding information, whether it's language, images, video, or audio.”
While that's pretty exciting — or slightly terrifying, depending on how you look at AI technology — AI technologies might be a bit further away.
Google says it's “working to incorporate the latest AI advancements into its products,” but that it's starting with a lightweight version in Google Search and Google Assistant.
However, in this article we have seen how the progress has been made in the field of Artificial Intelligence.
