Google's search engine has been the primary gateway to the internet for almost two decades, but in November 2022, it began facing its first potential competitor, the interactive chatbot ChatGPT, which has revolutionized the field of artificial intelligence and become the talk of the tech industry.
Through the robot ChatGPT, backed by Microsoft, the company OpenAI dominates the generative artificial intelligence market, having gathered more than 100 million active monthly users in just two and a half months since its launch, making it the fastest-growing web application ever. This significant success has pushed all technology companies to rush to provide updates and new products to keep up with this new technological revolution that we are currently witnessing.
So, what is Google Bard, how can you use it, and what are its features? Here's everything you need to know about this new competitor in the artificial intelligence market.
Google Bard is an interactive chatbot that operates on Google's AI and can respond to various queries and requests in a conversational manner. It is supposed to work similarly to the ChatGPT robot, with the main difference being that Google's robot will pull its information from the web.
Bard will not only be a chatbot, but Google is expected to integrate its capabilities into its search engine to provide a summarized answer to what the user is searching for, in addition to the traditional search results in Google's engine, although it is not widely available for use yet.
When did Google announce Bard?
Google announced Bard on February 6 in a statement issued by Sundar Pichai, the CEO of the company, which was published on the Google blog. Pichai confirmed that Bard will not just be a chatbot, but its capabilities will be integrated into the Google search engine.
Pichai also discussed the diversity of information that Bard can provide, ranging from simplifying complex scientific concepts to providing the latest sports news or teaching users new skills. He emphasized that Bard will not be limited to serving the average user, but Google will provide an API interface for developers to take advantage of its features and integrate them into their applications.
What is LaMDA?
Google Bard relies on LaMDA, the large language model developed by Google, which was revealed two years ago and is similar to GPT-3, the language model used in ChatGPT robot.
LaMDA differs from other language models in that it is trained on dialogue rather than text. Therefore, while GPT-3 focuses on creating text, LaMDA focuses on creating dialogue.
What makes LaMDA a remarkable achievement is that it can create conversations in a free, unconstrained manner that is not limited by task-based response parameters. The model uses concepts such as user intent, reinforcement learning, and recommendations so that the conversation can smoothly navigate between unrelated topics.
Google developed the LaMDA model based on Transformer, an open-source neural network architecture for natural language understanding announced by Google in 2017. It is one of the latest and most powerful categories of machine learning models invented so far. It is the cornerstone of many generative artificial intelligence applications that we see today, including the GPT-3 model used in ChatGPT.
In August 2021, researchers at Stanford University called the Transformer architecture "foundation models" in a research paper. These are flexible and reusable AI models that can be applied to almost any industrial field or task. They believe that foundation models represent a qualitative leap in artificial intelligence, as they have expanded our imagination in what is possible over the past few years.
Google describes Transformer as follows: "This architecture produces a model that can be trained to read many words, such as a sentence or paragraph, and pays attention to how these words are related to each other, then expects the words that it thinks will come next."
Therefore, the LaMDA model understands conversations because it is trained to understand the context of the dialogue. This allows it to keep up with the flow of the conversation and give the feeling that it is listening and responding accurately to what is being said.
The initial version of Bard will use a lightweight prototype of LaMDA, as it consumes less computer power, which helps Google offer the service to a larger number of users to get...
How to use Google Bard?
Google has not yet released Bard to the general public, as it is currently being tested with a small group of trusted testers, with plans to open it to all users in the coming weeks.
In this regard, Pichai said, "We will gather feedback from internal testing and user feedback from those testing Bard to develop the service and improve responses, to ensure that it is ready for the public while adhering to Google's responsible AI standards.
The Most Important Features and Uses of Google Bard
Robotic chatbots cannot currently conduct complete conversations like humans, but they do so by creating a wide range of digital texts that can be reused in almost any context, some of which are trained to understand text, and others are trained to understand dialogue context. Google's Bard robot aims to do this by:
◉ Collecting information from user responses and the web to provide new, high-quality responses.
◉ Using a lightweight prototype version of the LaMDA linguistic model for initial testing, which will evolve in future versions, meaning faster responses and more accurate answers.
◉ Collecting feedback to improve the artificial intelligence system in the future.
◉ Google also aims to revolutionize the information collection process and how we use search engines by developing its own artificial intelligence technology. Therefore, many use cases are expected to emerge for Google's Bard robot after its official launch, the most likely of which are:
Getting appropriate answers to queries using advanced artificial intelligence algorithms, as you can ask it about anything you want.
◉ Finding information more smoothly when using Google's search engine, whether you're looking for a weather report or the latest sports results, as it can help you with that.
◉ Simplifying complex information and explaining new discoveries.
◉ Easily automating complex tasks.
◉ Facilitating daily tasks, especially for activities such as time management and scheduling.
◉ Facilitating conversations with users on all topics.
Google has a long history of using artificial intelligence to improve search, and now uses the latest artificial intelligence technologies to develop entirely new ways to interact with information, whether through language, images, sound, or video.
Why is there a lot of controversy surrounding Bard right now?
A lot of controversy arose around Google Bard after Google announced it, to the point that reports indicate that Google lost $100 billion of its market value with the close of markets last Wednesday, despite it still being in the testing phase, and has not yet been released to the general public, so why?
This happened because of an incorrect information presented by the robot (Bard) in the promotional video that Google published during its press conference where it announced Bard.
When the robot was asked, "What are the new discoveries from the James Webb telescope that I can tell my 9-year-old son about?", the answer came in simplified bullet points, as the question requested, but it contained an error where one of the points stated that (James Webb) took the first picture of a planet outside of our solar system.
However, people quickly caught this mistake and shared tweets expressing their displeasure with Bard because the first picture taken of a planet outside of our solar system was in 2004, by the VLT (Very Large Telescope) belonging to the European Observatory, and this information is well-known and documented on NASA's website.
To respond to this, a Google spokesperson said to the media, "This highlights the importance of rigorous testing, which is what we have started doing with our Trusted Tester program," referring to the company's program that allows a group of users to try out its new services and provide feedback on them.
But before announcing Bard, the LaMDA model faced a lot of criticism, after Blake Lemoine, an artificial intelligence engineer at Google, presented a document to the company urging it to consider that the LaMDA model has consciousness and that it must fulfill its requirements.
Blake Lemoine said in June 2022 to The Washington Post that the LaMDA model has developed consciousness and emotional awareness, as it has the ability to perceive and express ideas and feelings equal to those of an eight-year-old child.
This controversy faded away after Google completely denied it, and gave Blake Lemoine the opportunity to explain his ideas further.
Why Did Google Decide to Reveal Bard Now?
As we mentioned before, the ChatGPT robot has achieved great success since its launch. A report released by the Swiss financial services group, USB, indicated that it is the fastest-growing web application ever. Due to this success, technology companies, including Google, are trying to secure their position in the rapidly evolving artificial intelligence market currently dominated by OpenAI, which is backed by Microsoft.
Moreover, this success has led Microsoft to increase its investments in OpenAI to reach $10 billion, and to gain the right to use ChatGPT in its programs. The day after Google's announcement of Bard, Microsoft announced the integration of an updated version of ChatGPT into its Bing search engine and Edge browser.
What other artificial intelligence services does Google possess?
Google has developed other artificial intelligence services that have not been released to the public yet. Google typically deals very carefully with AI products and does not release them until it trusts their performance. For example, it has developed an AI-based tool for generating images from text called Imagen, which is expected to be a great alternative to OpenAI's DALL-E tool when it is released.
Google has also developed a tool for generating music from text called MusicLM, but it says it has no plans to release it at this stage.
