Bard’s days are numbered. It was presented only a year ago in a rush in response to the emergence of ChatGPT, from OpenAI, to offer an automatic conversational response alternative. But it was only a model fresh out of the Google Labs, where the system that aspires to be definitive continued to develop: Gemini. This Thursday, the multinational announced that Bard was becoming Gemini and that it was available in the Web and in mobile applications with systems Android and also with iOS (Manzana). Initially, it only has an English version and will be expanded to Asian languages (first Japanese and Korean) and, later, Spanish. Paid versions will allow its use in common work tools such as documents or emails.
For years, Google has dominated the Internet search market. But OpenAI disrupted the system by developing an artificial intelligence model, ChatGPT, which was no longer content with providing websites that included, more or less correctly, search-related information. The new bot (an automated program with responses to specific instructions that mimic the human brain) was able to summarize and express its own solutions, an ability that quickly found its way into everything from school work to programming, from daily queries to industrial processes. .
The threat to Google’s hegemonic market led to a hasty reaction and Bard was presented with the temporary brand a year ago. This Thursday, this was confirmed and the model now becomes Gemini. The new system aims to be not only a replacement for the search engine, but also, according to Sissie Hsiao, vice president of Google, “a true artificial intelligence assistant, conversational, multimodal and more useful than ever.”
The company’s directive summarizes that the new application provides answers to written or spoken questions and also interacts with images. It will become the main assistant of the phone, accessible via a new shortcut or by the well-known voice command (Hi Google) or by terminal shortcuts. Conversations can be held and resumed from the same mobile phone or from the desktop computer.
“You can generate a headline based on the photo you took or reading an article and the app allows you to not only understand the article in more depth but also ask questions about it. Many Google Voice features will be available through the Gemini app, including setting timers, making calls, and controlling smart home devices,” says Hsiao.
These applications will be available to Google users, but at the time of everything is free has ended and the company intends to monetize the most comprehensive systems from the start. Thus, according to Sundar Pichai, head of the company, the version Advanced Gemini, which will be available by subscription, “is better able to reason, follow instructions, code, and collaborate creatively.” “For example,” he adds, “it can be your personal tutor and adapt to your learning style. They can also be a creative partner and help you plan a content strategy or create a business plan.
This advanced version, as it was announced during its presentation and received by Pichai, “utilizes a combination of 57 materials, including mathematics, physics, history, history, medicine and science, with the knowledge that it has so much knowledge as the capacity of problem resolution”.
Gemini Advanced will be available through the new Google One AI Premium plan and, with the subscription, compatible with up to five users, it can be used with Gmail, Docs, Sheets, Slides and Meet. Includes technical support, professional image editing, recommendations and early access to trial features. The initial offer in the United States involves, for existing Premium subscribers, a fee of 10 dollars (9.28 euros) more per month.
The Duet AI version, aimed more at businesses and developers, will become Gemini for Google Workspace and Google Cloud. The latter model has specific skills in process management, program development and protection against cyberattacks, according to the company.
Google’s commercial maneuver responds to OpenAI’s intention to launch the fifth version of ChatGPT this year, according to the company’s director, Sam Altman, who previewed some of its features. in the podcast by Bill Gates.
The bot aims to advance in terms of precision, personalization and multimodal capabilities, so that interaction will be with all types of text, digital and audiovisual files, separately or together.
In this way, ChatGPT5 aims to offer a more fluid and natural conversation, with capabilities to understand the user’s context, provide more creative capabilities and gain security against possible malicious uses.
To achieve this, language models were extended to training and programmed to learn from user interaction. Among its most common uses, the company anticipates that it will provide personalized care services, be integrated as an educational or entertainment tool, and be integrated into business management tasks.