Google has postponed the launch of its much-awaited conversational artificial intelligence (AI) dubbed Gemini, which is expected to compete with OpenAI's widely popular AI bot, ChatGPT.
To those unaware, Google announced Gemini at I/O 2023. Earlier this year, Google representatives told some cloud customers and business partners they would get access to the new conversational AI by November.
However, two people with direct knowledge reportedly told them not to expect Gemini's arrival until the first quarter of next year. A new report by The Information now suggests the search giant was originally slated to unveil its new foundation model next week with a series of events in California, New York and Washington.
Google Gemini struggling with non-English queries
According to the report, these impending events were aimed at politicians and policymakers. Gemini's release was reportedly postponed after Google "found the AI didn't reliably handle some non-English queries".
Google CEO Sundar Pichai stressed the importance of global language support, especially since Google is aiming to match or outperform OpenAI's GPT-4. According to sources, the company has "met that standard in some respects".
To recap, Pichai said in November that Google is "focused on getting Gemini 1.0 out as soon as possible, make sure it's competitive, state of the art, and we'll build from there on". Currently, the company is reportedly still "finalising the primary, biggest version of Gemini".
Can Gemini outperform GPT-4?
If rumours making the rounds online are anything to go by, Gemini can handle several types of data. Moreover, the model is reportedly capable of comprehending and generating text, images and other types of content including websites, based on written descriptions and sketches.
The word on the street is that Gemini exhibits superior performance over OpenAI's GPT-4 as it uses more computing power than its rival. While Google already has a generative AI model called Bard, it isn't as popular as ChatGPT.
However, some analysts believe this will change once Google launches its Gemini AI. Regrettably, the company is still mum on its plan to integrate Gemini into its services like Bard, Search and Workspace.
Nevertheless, earlier reports imply Google will offer Gemini in various sizes including a lightweight "Gecko" for mobile devices. Also, the company said Gemini will "enable future innovations, like memory and planning".
During I/O, the American tech giant noted that Gemini will get impressive multimodal capabilities that are not available in prior models. In addition to understanding text and images, the AI model will be highly efficient when it comes to tool and API integrations. Google is reportedly also testing AI assistants that can offer life advice.
