TechNews Pictorial PriceGrabber Video Tue Sep 19 22:00:54 2017

PopYard-->industries


Can Qualcomm Become a Big Name in Artificial Intelligence?
Source: Harsh Chauhan




It is absolutely essential for Qualcomm (NASDAQ:QCOM) to quickly find ways to tap emerging technologies in light of the challenges that it is facing. The chipmaker's licensing business has been down in the dumps ever since Apple started questioning its royalty structure.

This has turned out to be a royal pain for Qualcomm, as its revenue in the recently reported third quarter fell 11% year over year, driven by a massive 42% drop in licensing revenue. What's even more alarming is that the chipmaker could lose more licensing revenue, so it needs to quickly diversify into more areas to get back on track.

Therefore, it wasn't surprising when Qualcomm announced the acquisition of artificial intelligence (AI) company Scyfer for an undisclosed amount last month. But will the chipmaker's latest acquisition help it stand out in a crowded space where rivals like NVIDIA (NASDAQ: NVDA) are already putting their AI expertise into real-world applications? Let's take a look.
Man holding a tablet that's projecting a representative AI brain.

Image source: Getty Images.
Qualcomm has a different take on AI

Qualcomm is approaching AI applications in a different manner than other technology companies. The company is developing on-device AI solutions by running algorithms at the source instead of the cloud. Qualcomm claims that this approach will result in numerous benefits that include "immediate response, enhanced reliability, increased privacy, and efficient use of network bandwidth."

Qualcomm's method could help reduce data overload in the cloud and lead to quicker decision-making by the AI algorithms in real-world situations. For instance, an autonomous car    could process some of the AI algorithms on the vehicle itself instead of sending everything to the cloud. This could make the car more efficient at driving itself.

Qualcomm has been working on AI technology for a decade, and its Scyfer acquisition has the potential to further the company's progress.
What does Scyfer bring to the table?

Scyfer, a University of Amsterdam-affiliated company, was set up a couple of years ago, but has made impressive progress in a short time. Scyfer has already built AI solutions across a wide range of industries, including healthcare, finance, and manufacturing.

More specifically, Scyfer's AI solutions have already been used for financial prediction and quality control in the manufacturing business. Qualcomm can now diversify its AI research into more areas. Additionally, the chipmaker will get access to the AI expertise of the people at Scyfer, particularly founder Max Welling, which is why some are referring to the acquisition as an "acqui-hire."

Welling is the research chair of machine learning at the University of Amsterdam and is also affiliated with the University of California Irvine and The Canadian Institute for Advanced Research (CIFAR). He has now been appointed the vice president of Qualcomm Technologies in Netherlands, according to his LinkedIn profile.

This could turn out to be a key appointment, as Welling is part of a number of research labs including AMLAB, DeltaLab, and QUVA. Under his guidance, DeltaLab has been given a grant of 3 million euros from leading automotive component maker Bosch, indicating that the industry already recognizes Welling's expertise in this field.

Qualcomm's AI ambitions could get a shot in the arm thanks to Scyfer's expertise across a various range of applications.
What about the competition?

Rival chipmakers including Intel and NVIDIA are ahead of Qualcomm in AI, as their solutions are already being used commercially. For instance, NVIDIA's AI-enabled DRIVE PX2 platform is powering self-driving features in Tesla and Audi cars. Additionally, the chipmaker has penned deals with PACCAR and Toyota, which could help it put its self-driving platforms in mass-market cars and trucks.

But NVIDIA's AI reach isn't limited to just automotive -- the company has also made progress in the healthcare space. Its DGX-1 deep-learning supercomputer is being used by the Massachusetts General Hospital given its ability to use AI to study a patient's condition using past data and test results.

Intel, for its part, has made a spate of AI-related acquisitions, targeting different areas such as the cloud, automotive, drones, and virtual reality.
Foolish takeaway

The good part is that Qualcomm is trying out a different approach to tap the AI opportunity. Qualcomm can make AI algorithms more efficient by running them at the source instead of sending everything to the cloud, though it remains to be seen whether the industry will buy into this concept or not.

Additionally, the recent Scyfer acquisition should enhance Qualcomm's AI expertise, but it has a lot of ground to cover if it wants to make AI a big business for itself given the progress that rivals such as NVIDIA and Intel have made already.



© 2017 PopYard - Technology for Today!| about us | privacy policy |