Alphabet shares on a high as Gemini AI model makes its debut

Android’s Nearby Share technology is demonstrated at Alphabet’s Google Android plaza booth during the Consumer Electronics Show (CES) in Las Vegas, Nevada on January 5, 2023.
Android’s Nearby Share technology is demonstrated at Alphabet’s Google Android plaza booth during the Consumer Electronics Show (CES) in Las Vegas, Nevada on January 5, 2023. Copyright AFP
By Indrabati Lahiri
Share this articleComments
Share this articleClose Button

Google’s parent company Alphabet’s shares rose sharply on Thursday, after it unveiled its new Gemini AI model.

ADVERTISEMENT

Alphabet shares soared on Thursday, as the company announced Gemini, its AI model, as an answer to OpenAi, backed by rival Microsoft. 

Shares jumped 5.31% by the time markets closed on Thursday, trading at $136.9 (€127.04) and enabling the company’s market value to soar by another $80 billion.

Gemini is Alphabet’s long-awaited chance to reclaim its AI-pioneer glory, which was somewhat overshadowed by OpenAI’s ChatGPT launch last November. 

According to the company, Gemini will be able to process several types of audio, video and text formats. It is also supposed to be faster than OpenAI’s latest ChatGPT model, GPT-4 Turbo.

More interestingly, it can also apparently identify magic tricks, especially those based on sleight of hand, as well as put forward art suggestions depending on users’ available colours and other materials.

Users can also choose how much processing power they would like, with Gemini coming in three versions. The largest among these will be mostly directed towards data centres, whereas the smallest will be ideal for mobile devices.

According to Eli Collins, vice president of Product at DeepMind, Gemini is far cheaper than previous Google DeepMind AI products, which were larger but more expensive.

This is one of the biggest science and engineering efforts we’ve undertaken as a company
Sundar Pichai
Alphabet CEO

Concerns around how expensive developing GenAI models can become, as well as how to develop them more efficiently down the line, have led to Alphabet opting for these three different versions of its AI model.

Gemini is expected to go beyond anything Alphabet has come up with so far. According to Alphabet CEO Sundar Pichai, this is “one of the biggest science and engineering efforts we’ve undertaken as a company”.

The new AI model is also expected to help Alphabet pull back some of the embarrassment recently caused by its Bard chatbot sharing wrong information in a marketing video. This had led to the company losing about $100 billion of market value at the time, some of which has now been recovered by Gemini’s launch.

The company has also announced that it will be merging some of Gemini’s technology with its Bard chatbot, in what seems to be an attempt to further enhance the latter.

This could potentially be the right time for Alphabet to launch its own AI model, as ChatGPT starts to lose its shine somewhat, amidst increasing user complaints of inaccuracies following updates to the system.

As such, if Gemini does turn out to be as good as Alphabet claims, there is potentially a wide open market to draw ChatGPT’s disgruntled customers. 

This could also be a good opportunity for Alphabet to claw back some of the AI-driven cloud revenue Google Cloud lost to Microsoft Azure. Google Cloud fell to three-year lows back in the third quarter of this year, while Azure’s market share soared.

Currently, Alphabet is targeting more startups, which already use its software, in an attempt to boost Gemini’s sales and awareness.

Share this articleComments

You might also like