May 24, 2024
Gemini AI

Gemini AI

Googles latest AI model beat Open AI’s chatbot in 30 out of 32 standard benchmark tests.

In the I/O developer conference CEO of the tech giant Sundar Pichai has declared the new era of ‘Gemini AI’ the latest AI model launched by Google, which may wipe out Chatgpt from the market.

Gemini’s coding ability is a breakthrough. It can code better than 85% of coders in coding competitions. Google got away with Bards failure with the launch of Gemini.

Gemini is much more than a model. Gemini contains advanced functions and it has versions to suit different purposes. A lighter version called Nano is specially designed to run on Android phones. The bulkier version known as Gemini Pro will be the backbone of Bard and will support Google AI services. The most capable version is known as Google Ultra, mainly designed for data centers and office applications.

Gemini AI could boost technology and business: But competitor`s are too close

Till now general public uses AI models for different uses and thus Gemini has been the most updated version of tech till now.

The “multimodality” feature of the chatgpt sets it apart from the competitors. We use Google products everyday and this AI transformation gave a boost in technology to the tech giant.
Gemini brings all the uses of AI models under one app hence simplifying and facilitating user experience.

Gemini certainly has its issues which will come in front as it will go through different phases of uses.

Gemini`s demo video came out unacceptable

Googles new revelation backfired after the impressive debut. The tech giants authenticity came under question after the demo video of Gemini is found fake. The video on how the multi-model works ( “Hands-on with Gemini: Interacting with multi-model AI”) hit a million views as readers were curious to watch the new hype. They were eager to understand that how the multi-model used mixed languages and is flexible and responsive to multiple inputs.

In the beginning, it shows an evolving sketch of a duck from a squiggle to a completed drawing, which it says is an unrealistic color, then gives a surprise (“What the quack!”) when seeing a toy blue duck. It then answers to various voice queries about that toy, then the demo moves on to other show-off moves, it tracks a ball in a cup-switching game, recognized shadow puppet gestures, reordered sketches of planets, and so on.

The video was mind-blowing but one problem is the video was fake. Google made it with a text prompt and image prompt.

“We created the demo by capturing footage in order to test Gemini’s capabilities on a wide range of challenges. Then we prompted Gemini using still image frames from the footage, and prompting via text.” (Parmy Olson at Bloomberg was the first to report the discrepancy).

“Google is beginning to address investor concerns around generative AI innovation and the high cost of running GenAI models through the combination of Gemini’s different model sizes,” J.P. Morgan analysts said.

After multiple requests for comment, the company on Friday told CNBC in a statement, “The video is an illustrative depiction of the possibilities of interacting with Gemini, based on real multimodal prompts and outputs from testing. We look forward to seeing what people create when access to Gemini Pro opens on December 13.”

About The Author

Leave a Reply

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

Copyright © All rights reserved. | ChromeNews by AF themes.