Google’s new Gemini app—launching in India today, with support for nine Indian languages no less—is a cutting-edge generative AI-powered mobile application designed to enhance productivity and streamline user interaction through advanced AI capabilities. In simple words, it brings Google generative AI in the palm of your hands.
Launched as a successor to Bard, Gemini aims to provide a more sophisticated and integrated AI experience across both Android and iOS platforms, though the dedicated app is only available for Android.
What is Google Gemini?
Gemini (previously Bard) represents a new era in AI technology, Google says. Developed by Google Deepmind, it is designed to function as a versatile assistant capable of understanding and responding to a wide range of inputs. Google doesn’t say it out loud but Gemini—aims to—eventually replace the older Google Assistant with a new bunch of capabilities built for the modern era that it speaks of.
Gemini is also the name of the large language model driving it. (For some comparison, OpenAI’s rival app ChatGPT is powered by GPT, the latest version being GPT-4). The LLM brings multimodal capabilities, meaning it can process and respond to text, images, and voice commands. This allows Gemini to answer both simple queries as well as analyse complex data and generate a response within no time. This is done using “tokens”, that LLMs use to generate and process language.
Key features and functions
Advanced Conversational AI: One of the core features of Gemini is its advanced conversational abilities. It is designed to engage in natural, context-aware dialogues, making it an effective virtual assistant for personal and professional use.
Integration with Google ecosystem: Gemini integrates seamlessly with Google’s suite of services, including Gmail, Google Drive, Google Calendar, and more. This integration allows users to perform complex actions with simple commands, enhancing workflow efficiency and productivity. For example, you can ask Gemini to find and summarise emails, schedule events, or analyse documents stored in Google Drive. Google is bringing Gemini to the Google Messages app as well, in English to begin with.
Personalisation and Adaptability: Through continuous interaction, a chatbot like Gemini can learn and adapt to individual user preferences and routines. This personalised approach ensures that the AI can provide more relevant suggestions, reminders, and insights over time. This means, the more you use Gemini, the better it understands your habits and needs, improving its overall utility.
Long Context Window: Google’s next-generation Gemini 1.5 Pro model features an extensive context window that can handle up to 1 million tokens (though it’s for paid users only under the Gemini Advanced experience). This allows the AI to process large documents and multiple inputs simultaneously, making it particularly useful for detailed analysis and summarisation tasks.
How to download and use Google Gemini
Using Google Gemini is straightforward and accessible across multiple devices:
- Android: Download the Gemini app from the Google Play Store. It is available in India now. You can launch it by tapping the app icon, long-pressing the power button, or simply using the wake word “Hey Google.”
- iOS: While there’s no dedicated Gemini app for iPhone, Gemini is integrated within the existing Google app. You can access it by switching to Gemini in the Google Discover feed.
A word of caution: Generative models while remarkable still have some limitations. They’re prone to hallucinations, for instance. They may produce incorrect, misleading, or nonsensical information some times, the reason being that the language models powering them generate responses based on patterns in the data they were trained on, instead of understanding the content in a way a human would. And so, they might create plausible-sounding but factually inaccurate or logically inconsistent responses. Google’s AI overviews feature itself came under the scanner for such hallucinations recently, suggesting cats have been to the moon or it was okay to use glue to put cheese on your pizza. That’s something you should always keep in mind while using these chatbots. They may be enablers, but nothing beats the human brain and intuition—at least not yet anyway.
Follow FE Tech Bytes on Twitter, Instagram, LinkedIn, Facebook.