Google’s much-anticipated next-gen AI model, Gemini, has faced setbacks, leading to a rescheduled launch in January 2024 instead of its imminent reveal.
Originally scheduled for a big unveiling next week, Gemini encountered hurdles in handling certain non-English queries, prompting the postponement of launch events in California, New York, and Washington, as reported by The Information. This delay comes after Google’s announcement of Gemini at I/O 2023 in May.
The delay primarily stems from Gemini’s struggle with non-English queries, emphasizing Google’s commitment to robust global language support. Despite these challenges, insiders note that Gemini exhibits superiority in some aspects compared to GPT-4, aiming to surpass OpenAI’s model with its advanced text and image generation capabilities.
At I/O 2023, Google showcased Gemini’s exceptional multimodal capabilities, a unique trait among existing models. The AI model adeptly handles text and images, seamlessly integrating various tools and APIs. Though still in finalization, Gemini’s potential is demonstrated in its ability to generate multi-format content based on prompts, as highlighted by Sissie Hsiao, Google’s vice-president.
Gemini’s competitive edge lies in its versatility with multiple data types, including text, images, and website content, outperforming GPT-4 through enhanced computing power. As anticipation builds for Gemini’s launch, it’s poised to reshape AI applications, potentially shifting consumer attention from current models like ChatGPT.
While its integration into Google services like Bard, Search, and Workspace remains uncertain, Google plans to offer diverse versions of Gemini, including a mobile-friendly “Gecko.” Moreover, Gemini’s design aims to lay the groundwork for future innovations, including memory and planning features.