Google has discreetly unveiled its latest AI model, Gemini 2.0 Pro Experimental, in an update to the changelog of its Gemini chatbot app. Unlike previous launches, the company chose not to make a major announcement, instead revealing the model’s arrival through a simple update.
However, the changelog mentioning the new model was later removed, with Google clarifying that an outdated release note was mistakenly published.
The new model succeeds Gemini 1.5 Pro and is available to Gemini Advanced users starting Thursday. Google claims it offers improved factual accuracy and enhanced performance in coding and mathematics.
It aims to assist with complex programming tasks, such as generating code from scratch, and solving advanced mathematical problems, including statistical models and quantum algorithms. Despite these advancements, Google acknowledges that Gemini 2.0 Pro Experimental is still in an early preview phase and may exhibit unexpected behavior or errors.
This launch comes as competition in the AI industry intensifies, particularly with Chinese AI startup DeepSeek gaining global attention.
DeepSeek’s latest models, which are openly available for companies to use, have been challenging the dominance of major U.S. tech firms by offering comparable or superior performance. This shift has prompted discussions within Silicon Valley and at government levels about the future of AI development.
The new AI model is accessible through Google’s Gemini for Google Workspace and the Google One AI Premium plan. However, it lacks access to real-time information and is incompatible with some existing Gemini app features.
As Google continues refining its AI capabilities, this model is expected to be an important step toward further improvements in AI-powered applications and tools.
