Google launches Gemini, its most powerful artificial intelligence; see how it works

[ad_1]

The model was integrated into Bard, ChatGPT’s rival, and will reach other Google products, such as Search and Chrome, in the coming months. AI was created to work both in large infrastructures, such as data centers, and on cell phones. Google launched this Wednesday (6) Gemini, an artificial intelligence (AI) model that is treated by the company as the most powerful ever created by its team. It can now be tested on Bard (ChatGPT’s competitor), but, for now, only in English. According to Google Gemini surpassed human ability in a knowledge and problem-solving test that combines 57 topics, including mathematics, physics and history. The company also says that it performed better in this test than GPT-4, the OpenAI model used in ChatGPT. Learn more about Google’s new artificial intelligence: What can Gemini do? How does Gemini work? When will it be available? What makes Gemini more powerful? Is Gemini reliable? What does Gemini do? According to Google, Gemini will be useful for carrying out tasks that require more reasoning capacity. It can be used both to help programmers with complex codes and students with their homework, for example. In a demonstration, Gemini was able to analyze a photo of a physics exercise and identify an error in the calculation. The AI ​​was able to present the correct solution and create similar problems for the user to try to solve. Google demonstration shows that Gemini analyzes physics exercises and identifies when there is a calculation error Disclosure/Google How does Gemini work? Gemini is the most flexible AI model ever developed by Google. The idea is that it can run both on large infrastructures, such as data centers, and on more limited devices, such as cell phones. To meet this objective, the new artificial intelligence will have three versions: ⚡ Gemini Ultra: larger and more powerful, aimed at highly complex tasks – will only be released in 2024; 💻 Gemini Pro: aimed at running a wide range of tasks and serving users, starting this Wednesday on Bard, and developers, starting December 13th; 📱 Gemini Nano: created for mobile devices, it will run directly on the device, which allows it to work even when there is no internet – available on the Pixel 8 Pro, Google’s cell phone, to create audio summaries and suggest intelligent responses on WhatsApp. Gemini, Google’s AI, was developed to be flexible and work from data centers to cell phones with Ultra, Pro and Nano versions Disclosure/Google R$ 24 thousand salary: the IT career that should boom in 2024 Taylor Swift, Barbie and Flamengo : what’s hot on Tinder this year Back to top When will it be available? Many users’ first contact with Gemini should be with the Pro version, which will improve reasoning, planning and understanding of Bard. Starting this Wednesday, the Google robot will already use the new AI in responses in English (other languages ​​will receive the update soon). In early 2024, Google will launch Bard Advanced, which will make Gemini Ultra available on Bard. And, according to the company, products such as Search and Chrome will integrate with Gemini in the coming months. Starting December 13, developers and businesses will be able to use Gemini Pro through the Google AI Studio and Google Cloud Vertex AI artificial intelligence platforms. And Gemini Ultra will undergo more security checks before being released to everyone in 2024. Bard can be used in Google search; see how to do it ChatGPT model is available on Bing, Microsoft’s search engine Back to top What makes Gemini more powerful? Google says that the approach to developing Gemini was a little different from that adopted in other multimodal artificial intelligence models, that is, that can understand different formats at the same time, such as texts, images, audio and codes. According to the company, the creation of AI models usually involves separate “training”, focusing on each format. In a second stage, text learning is brought together with image learning, for example, and everything is “packaged” into just one product. In Google’s assessment, this method allows you to perform some tasks, such as describing images, but has limitations in cases that require more reasoning. “We designed Gemini to be natively multimodal, pre-trained from the ground up on different modalities. We then refined it with additional multimodal data to further refine its effectiveness,” said Demis Hassabis, CEO of DeepMind, Google’s AI subsidiary . How to remove ‘online’ from Instagram and use the app without anyone knowing The Brazilian supercomputer among the fastest in the world Back to top Is Gemini reliable? Google admits that, like all artificial intelligence, Gemini is also subject to the problem of hallucination, an error that causes a robot’s response to include incorrect, biased or meaningless information. But the company says Gemini is capable of making the best security assessments of an AI its team has ever created. The objective is for it to be able to avoid violent or stereotypical content, for example. The creation of Gemini also involved training on what should not be displayed to users based on a set of 100,000 phrases considered toxic taken from the internet. A group of independent experts is also working to test the limits of new AI. “We are approaching this work boldly and responsibly,” said Sundar Pichai, chief executive of Google parent Alphabet. “This means being ambitious in our research and pursuing capabilities that will bring enormous benefits to people and society, while building safeguards and working collaboratively with governments and experts to address risks as AI becomes more capable.” “, he stated. READ ALSO: Can anyone who posts memes with photos of third parties be sued? Understand what the Cybertruck law says: what the almost ‘unbreakable’ car that Tesla sells in the USA is like Elon Musk attacks advertisers after Twitter boycott: ‘Go f…’ Back to top Bard, ‘Google’s ChatGPT’ , is launched in Brazil

[ad_2]

Source link