Google launches Gemini 3 to challenge ChatGPT and announces a focus on rigor with GPT-5
Google has introduced a new generation of its AI, Gemini 3, in a clear move to compete with OpenAI’s ChatGPT. The company highlights that this model brings significant improvements in reasoning, multimodality (text, image, audio, video), and more powerful programming agents.
At the same time, Google states that it intends to maintain high standards of “rigor” — suggesting that the advancement of OpenAI’s GPT-5 will be monitored more closely, in a kind of “responsibility race” between the AI giants.
According to analysts, the launch of Gemini 3 reinforces Google’s strategy of heavily investing in generative AI integrated across its products (such as Search), while also responding to criticism regarding inaccuracies and excessive “flattery” in earlier AIs.
For Google, Gemini 3 represents a decisive step toward a more reliable and powerful model, strengthening competition with OpenAI in a landscape of increasingly sophisticated AI systems.
Beyond simple text
One of the system’s main innovations is the fact that it is “natively multimodal.” In simple terms, this means the artificial intelligence (AI) can process text, images, and audio simultaneously, rather than dealing with each element separately. This approach enables a more fluid and natural interaction. To illustrate, Google offers a practical example: the user can photograph a handwritten recipe, and Gemini 3 Pro transforms those images into an organized cookbook; or, alternatively, it creates interactive study cards from a set of video lessons.
According to Doshi, the advancement consists of “moving beyond the paradigm of text-only responses” and offering a richer and more complete experience. The change will be noticeable in the so-called “generative interfaces”: in internal tests, the system is capable of producing visual layouts similar to a magazine, with navigable images, or generating dynamic diagrams tailored to the user’s request.
CNBC
0 Comments