Google delays launch of Gemini, its ChatGPT rival. What you need to know

Google’s new AI Chatbot, Gemini, was supposed to launch this week. However, its release has been postponed due to the chatbot struggling to handle some non-English queries properly, according to a report.

According to sources from The Information, Sundar Pichai, the CEO of Google, has decided to cancel a series of planned Gemini events scheduled for next week in California, New York, and Washington.

The delay in the chatbot's launch is attributed to Gemini's inconsistent effectiveness in dealing with queries in languages other than English. Recent reports suggest it might now launch in January 2024.

Google's conversational artificial intelligence 'Gemini' has been developed to compete with OpenAI’s ChatGPT.

Gemini's purpose

Google aimed to create Gemini to handle various applications, including integrating different data types such as text, images, and code. Anticipated to use multimodal learning techniques, this AI is expected to process and generate various data formats like text, images, code, and audio.

It is widely believed that Gemini will enhance Google's existing artificial intelligence and AI-powered products, including Bard, Google Assistant, and Search.

Will Gemini outperform ChatGPT upon launch?

Demis Hassabis, CEO of DeepMind, asserted that Gemini would surpass ChatGPT's capabilities upon launch, stating, "When Gemini is complete, it could play a major role in Google’s response to the competitive threat posed by ChatGPT and other generative AI technology."

AI Programme used to design Gemini: Demis Hassabis highlighted that engineers are using techniques from the Artificial Intelligence program called AlphaGo to develop Gemini.

How is it different from Google Bard?

Earlier this year, Google launched a conversational generative artificial intelligence chatbot named 'Bard.' Initially criticized for making factual errors, these issues were later rectified, and the chatbot's responses improved. Google is now preparing to launch its language model 'Gemini' next year.

Gemini is expected to exhibit strong generative capabilities, enabling it to produce human-quality text, translate languages, create content, and more.

Target audience

Bard: Geared towards general users seeking information and engaging in open-ended conversations.

Gemini: Potentially targets professionals and developers who need an AI tool for tasks like creative content generation, code translation, or data analysis across different formats.

Technical approach

Bard: Trained on a massive dataset of text and code, using Transformer architecture.

Gemini: Leverages techniques from AlphaGo, incorporating multimodal learning to process and generate various data formats. This might allow it to understand and respond to non-linguistic cues better.