Shares in computer chip designer Nvidia have soared over the past week, taking the company’s valuation above the one trillion dollar mark.
GETTY IMAGES.
Nvidia was known for its graphics processing computer chips.
That’s what being a trillion dollar company means – it connects with Apple, Amazon, Alphabet, and Microsoft in one club.
This happened following the announcement of its second quarter earnings which came late in the night on Wednesday. The company’s spokesperson said that its chip production was being raised in response “to surging demand”.
Today, Nvidia leads this sector of selling AI chips.
That area of interest became a madness after ChatGPT became available publicly for the first time, spreading even beyond the field of technology.
Starting from speech assistance and moving on through software coding and cookery, it is clear that ChatGPT is by far one of the most widespread applications of AI.
The company has added itself to a unique club of 5 US billion dollar companies such as technology giants Apple, Amazon, Alphabet, and Microsoft.
This happened after it announced its most recent quarterly figures that came out very late on a Wednesday. The corporation stated that it would increase chip manufacturing in light of “the rising demand”.
Now, Nvidia rules the market of artificial intelligence (AI) chips.
After release of ChatGBT last time November, interest for that sector reached madness and that was beyond only technologists.
Chat GPT, one of the new applications of AI and one that is extremely popular, assists in giving speeches, coding, cooking and others.
However, such could only happen if some strong computer hardware existed — in particular Nvidia chips originating from California.
Most AI applications are supported by Nviidia hardware which was originally known as a company specialized in manufacturing graphics processing units for computer games.
“Says Alan Priestley, a semiconductor industry analyst at Gartner, “This is the leading technology player for the new thing called artificial intelligence”.
According to Dan Hutcheson, a TechInsights’ analyst, “NVIDIA is to AI as Intel was to PCs.”
ChatGPT is a product that was taught utilizing 10,000 GPUs that were combined together and connected to a supercomputer, owned by Microsoft
According to Ian Buck, general manager and vice president of accelerated computing at Nvidia, “IT IS ONE OF MANY SUPERCOMPUTERS – SOME PUBLIC AND SOME NOT – THAT WERE BUILT USING NVDIA GPUS FOR A RAN
The recent research made by CB Insight noted that Nvidia holds about 95% GPU market share dedicated for machine learning.
The chips are approximately of $10,000 (£8,000), but its newest and most powerful models sell at much higher prices.
Now, where was Nvidia that it became the most important player on the AI revolution?
In brief, a wager with courage on its own technology and a pinch of luck in timing.
It is one of many such supercomputers; some public and private that are equipped with Nvidia GPUs for both scientific and AI purpose.
According to one recent report by CB Insights, over 95 percent of the GPU market for machine learning is controlled by Nvidia.
The company’s AI chips cost approximately $10,000 (£8,000). However, the latest and the most powerful version is much higher in price.
Then, what made Nvidia the leader of an AI revolution?
To sum up; bold betting of self-technology and little bit of time luckiness.
One of its founders was Jensen Huang who is currently the CEO of nvidiaback in 1903. Secondly, Nvidia’s interest was in improving graphics for games and other purposes.
It came up with GPUs in 1999 to enhance computer’s image display.
GPUs are excellent when it comes to processing lots of these small jobs, otherwise known as parallel processing, which includes millions of pixels in this current instance.
Researchers from Stanford University found out in 2006 that aside from the gaming purpose, the GPUs can also be utilized as a means of speeding up the math’s functions which are beyond the capability of the common processing chip.
At that point, Mr Huang made a critical move towards what AI has attained now.
Instead, he used his company’s resources into designing a programming tool that unleashed the potential of GPUs not only for graphics but also for other applications that take advantage of parallel processing.
The company was among various other companies that were in the process of making their respective computer chip. Players only did not require this capacity and might not have known that they had it while the researchers offered another form of high performance computing at consumer hardware.
Those were those capabilities responsible for triggering initial successes of contemporary AI.
Alexnet, an AI that can classify images, was introduced in 2012. In training Alexnet, only two of Nvidia’s programmable GPUs were utilized.
Instead of months that it might take for hundreds of the typical processing chips, the training process lasted only a few days.
Word also quickly spread out to the computer scientists who began purchasing these cards in order to try applying them for running a brand new class of job.
According to Mr Buck, “We were caught by AI”.
Nvidia made an added advantage through pouring investment on different types of GPUs specialized for AI and other related softer wares which simplify its use.
Ten years down the line plus billions of dollars in the making, ChatGPT was born- an artificial intelligent entity so human sounding in its responses.
Deep fakes of Tom Cruise by Metaphysic in 2021.
Photorealistic videos of celebrities and others are produced by an American AI start-up named Metaphysic. The Tom Cruise deep fakes it made that year had everyone talking about it.
It runs and trains its models on hundreds of Nvidia GPUs; some bought at NVIDIA and other acquired via cloud computing services.
According to Tom Graham, its co-founder and CEO, “Nvidia is the only option out there that we could use.” This is “way out in front”.
Nonetheless while Nivdia seems to be safe at least in short time span, the long time prediction is hard to make. According to Kevin Krewell another industry analyst at TIRIAS research says, “Nvidia has a bull’s eye on it that every competitor is aiming to shoot.”
Some competition comes from other major semiconductor companies. Although, other two companies also know much as CPU manufacture. AMD is a well-known producer, making dedicated GPUs for AI purposes (Intel entered this market pretty lately).
Google’s TPUs are employed for both results on search and some specific machine learning functions, whereas one for the training of AI models in Amazon’s case.
In addition, there are rumors of a Microsoft’s AI chip, as well as Meta’s own AI chip development project.
Leave a Reply