Google says new AI model Gemini outperforms ChatGPT in most tests

Google has introduced a groundbreaking artificial intelligence model named Gemini, boasting superior performance compared to ChatGPT in various tests and demonstrating “advanced reasoning” capabilities across diverse formats. This unveiling comes on the heels of last month’s global AI safety summit, where tech companies pledged to collaborate with governments to test advanced systems before and after release. Google has indicated ongoing discussions with the UK’s newly established AI Safety Institute for testing the most potent version of Gemini, slated for release next year.
According to Google, the Ultra version of Gemini has outperformed “state-of-the-art” AI models, including ChatGPT’s formidable GPT-4, across 30 out of 32 benchmark tests, showcasing reasoning and image understanding prowess. The Pro model, another variant of Gemini, surpasses GPT-3.5, the foundational technology for the free-access version of ChatGPT, in six out of eight tests.
Gemini is a multimodal model capable of simultaneously comprehending text, audio, images, video, and computer code. Google plans to integrate Gemini into its products, including the search engine, with an initial release in over 170 countries, including the United States, on Wednesday. The launch will be an upgrade to Google’s chatbot, Bard.
However, Google will not initially roll out the Bard upgrade in the UK and Europe as Google awaits regulatory clearance. This cautious approach aligns with the company’s commitment to compliance and aligning with regional regulations as it introduces this advanced AI model to users worldwide.

Source: The Guardian



, ,



Leave a Reply

Your email address will not be published. Required fields are marked *