Earlier this week, the Softbank Group sold chip design company Arm Ltd. to Nvidia for 40 billion USD in stocks and cash. The deal is expected to take up to 18 months to complete, however it has already been fully approved by Arm, Nvidia and Softbank. Arm holds the title of the most prominent mobile chip provider, and now that America’s largest chip maker owns Arm, this is the biggest deal that has ever been made in the industry. Nvidia aims to strengthen their position as leaders in artificial intelligence computing with this deal.
What is the reasoning behind the deal?
Arm, which is based in the UK, provides the blueprints used in more than 90% of smartphone chips, which means that all major cellphone manufacturing companies such as Apple, Samsung, Huawei and many more rely on this technology for the design of their chips. On top of this, the company has plans to move into the data centre server and PC market, areas that are currently dominated by Intel. As it stands, Arm based server CPUs only hold a market share of 1-2% in comparison to Intel’s 95%. Nvidia hopes that by combining Arm’s energy efficient CPUs with the power of Nvidia’s graphics processing units, they can offer an alternative to Intel.
The deal also stands to solidify Nvidia’s position as leaders in AI computing, powered by their popular graphics processing units. Powerful super computers can be made by combining the capabilities of Nvidia and Arm tech.
Possible Downsides to the deal.
One of the main issues that might arise is that UK based Arm has always been neutral by licensing chip blueprints to the chip developers, charging an initial fee and later collecting royalties if successful chips were produced. There are worries that this neutrality will be lost in the hands of Nvidia, as Arm has access to chip manufacturer’s confidential chip shipment plans. When softbank acquired Arm in 2016, there was no such worry as Softbank did not directly compete with Arm’s clients. While Nvidia has been out of the mobile chip market since 2014, the company’s chip design capabilities mean that it could move into markets that compete with Arm clients, such as server chips.
The CEO of Arm assured the public that there will be no loss of neutrality, saying that “Independence is part of the value of Arm. … We may have a different parent company, but we will continue to keep the level of independence without changing the business model.” Jensen Huang, the CEO of Nvidia, had a similar remark, saying that “We will protect its business model, keep its openness, keep it neutral and let people keep building on Arm with benefits from our technologies as well.”
The future of AI.
Jensen said that AI is the most powerful technology of our time. By working with Arm, Nvidia looks to expand the use of AI and strengthen the architecture behind it to allow for more progress. Jensen went on to say, “AI is moving from the cloud to the edge, where smart sensors connected to AI computers can speed checkouts, direct forklifts, orchestrate traffic, save power. In time, there will be trillions of these small autonomous computers, powered by AI, connected to massively powerful cloud data centers in every corner of the world.” A big issue with AI is power consumption, and the deal between Nvidia and Arm can overcome this issue, resulting in more applications of AI at less expense. Nvidia can boost the power efficiency of Arm CPUs, bringing down this power consumption, while Arm can hugely benefit Nvidia’s AI expertise and resources.
While we do have to wait for the deal to be finalized, we can expect exciting progress being made from both Arm and Nvidia in the future, with both companies working together to improve their products.