Google Launches Advanced Chatbot ‘Bard’ Powered by Gemini AI

In a continued effort to compete with OpenAI’s ChatGPT, Google has unveiled an upgraded version of its chatbot named Bard. This new release, backed by Google’s latest AI technology called Gemini, is designed to deliver more accurate responses and emulate human reasoning more closely, according to Sundar Pichai, Google’s CEO. The updated chatbot is available for English speakers in over 170 territories, including the United States.

Pichai stated, “This marks the beginning of the Gemini era,” highlighting the company’s commitment to integrating three different versions of the technology into various products and services in the coming months. Gemini has been under development since the start of the year and is poised to enhance Google’s capabilities in the AI space significantly.

When OpenAI introduced the groundbreaking ChatGPT, Google faced a challenge to match its capabilities. In response, Google launched its initial chatbot, Bard, in March, which received mixed reviews. Subsequently, Google merged its two AI labs, Google Brain and DeepMind, and announced the development of Gemini at its Google I/O conference in May.

Gemini, described as more powerful than Google’s previous chatbot technologies, demonstrates improved performance, generating accurate responses and displaying advanced reasoning capabilities in certain scenarios. The technology is set to be integrated into various products and services, marking a significant step in Google’s AI strategy.

Gemini’s benchmark test results, released by Google, indicate that its most powerful version outperforms OpenAI’s GPT-4 in key areas. Notably, it excels in generating computer code and provides more accurate summarizations of news articles and other text documents. While Gemini is designed to analyze images and sounds, these features will be added to the Bard chatbot at a later date.

Google has developed three versions of Gemini with different sets of skills: Ultra, Pro, and Nano. Ultra, the largest version, is designed for complex tasks and is expected to debut next year. Pro, the mid-tier offering, will be integrated into various Google services, starting with the Bard chatbot. Nano, the smallest version, will power specific features on the Pixel 8 Pro smartphone.

Gemini represents a large language model (L.L.M.), a complex mathematical system capable of learning skills by analyzing extensive data sources, including digital books, Wikipedia articles, and online forums. This multimodal system enables Gemini to analyze and respond to both images and sounds, making it more versatile in various applications.

Despite its advancements, Google acknowledges that Gemini, like other similar systems, is not flawless and may make mistakes, including factual errors or “hallucinations” where it generates inaccurate information. Google Cloud, the company’s cloud computing services arm, is keen to offer Gemini to clients, competing against OpenAI and Microsoft for deals. The company has created a migration plan to attract OpenAI customers, offering comparable rates and additional cloud credits.

Google Cloud clients will gain access to Gemini Pro on December 13, while testing of Gemini Ultra, the most powerful version, is already underway. Despite the intense competition with OpenAI, Pichai emphasized that there is enough room for multiple AI providers in the market, signaling the industry’s early stages of development.

Pichai expressed excitement about Google’s progress, acknowledging that the company is still in the early days of AI innovation. He stated, “We have a sense of excitement at what we’re launching.

Google’s release of the Bard chatbot powered by Gemini marks a significant step forward in the company’s AI endeavors, positioning it as a major player in the competitive landscape alongside other industry leaders. The multifaceted capabilities of Gemini and its integration into various products and services underscore Google’s commitment to advancing AI technology.