Google DeepMind, a pioneer in AI research, announced Gemini, a groundbreaking multimodal AI model that can handle various types of information, such as text, code, audio, image, and video. Gemini is the result of a collaborative effort between DeepMind and other Google teams, and it aims to revolutionize the field of AI applications.
According to Demis Hassabis, the CEO and co-founder of Google DeepMind, Gemini is “the most capable and general model we’ve ever built.” Gemini can perform sophisticated reasoning tasks across different domains, surpassing human expert performance on the Massive Multitask Language Understanding (MMLU) benchmark. Gemini can also understand and generate high-quality code in popular programming languages, outperforming its predecessor, AlphaCode, in competitive programming.
Google CEO Sundar Pichai expressed his excitement for the potential of Gemini, saying that it will create unprecedented opportunities for people worldwide. “The transition we are witnessing with AI will be the most profound in our lifetimes, far surpassing previous technological shifts,” he said. “Gemini has the power to foster innovation and economic progress on an unprecedented scale.”
Google is making Gemini available to the public through its products and platforms. Starting with Bard and Pixel, Gemini will roll out across more Google products, such as Search, Ads, Chrome, and Duet AI, in the coming months. Developers can also access Gemini through Google AI Studio or Google Cloud Vertex AI.
Gemini 1.0 comes in three different sizes: Ultra, Pro, and Nano, optimized for different use cases and scenarios. It can extract insights from vast amounts of data, leading to breakthroughs in fields ranging from science to finance.
Google is also committed to responsible AI development, ensuring that Gemini meets the highest standards of safety and security. The AI undergoes rigorous testing and evaluation, including checks for bias and toxicity, as well as potential risks in areas like cyber offense, persuasion, and autonomy. Google is also collaborating with the rest of the tech industry to define best practices and guidelines for AI development.
Gemini is just the beginning of a new era of AI. It promises to enhance creativity, extend knowledge, and transform the way billions of people around the world live and work every day.
For more information about this news, visit this blogpost.
