At the recent SIGGRAPH conference, Meta head Mark Zuckerberg announced that people would be able to create digital twins of themselves using the newly launched Meta AI Studio.
“Every creator can build an AI version of themselves as an agent or an assistant that their community can interact with. There’s a fundamental issue here that there are just not enough hours in the day, if you’re a creator who wants to engage more with your community,” Zuckerberg said, speaking to NVIDIA CEO Jensen Huang.
However, this isn’t the only way the digital twin technology is being used. Looking at the enterprise side of things, companies have begun releasing the use of digital twins in testing out systems, especially those that are risky to implement on a large-scale, like AI integration.
Speaking to AIM, Keysight Technologies country general manager and VP Sudhir Tangri said that digital twin technology will be the centrepiece when it comes to AI integration and 6G research.
This California-based communications and electronics testing company partnered with NVIDIA in April this year, launching the 6G Research Cloud Platform to allow researchers to help advance wireless technology, particularly 6G.
This partnership involved several key players apart from Keysight, including Samsung, Ansys and Nokia, leveraging their capabilities to build the platform to make use of AI in advancing research for 6G technology.
A part of the cloud platform includes the establishment of the NVIDIA Aerial Omniverse Digital Twin for 6G. The digital twin is able to create accurate simulations of entire 6G systems, visualising anything from a single tower to an entire city network.
“So we are very actively pursuing those areas. We have joined the AI alliance with NVIDIA, SoftBank and all of the other big companies to create real solutions for AI in the radio access network,” said Tangri, referring to the Aerial Omniverse Digital Twin.
What Makes Digital Twins So Good For Testing?
Digital twin technology isn’t just used to test out 6G technology. With thousands of companies switching to the cloud, testing these systems remains crucial, but going about testing virtual systems in a way that is risk-free, yet accurate, remains a challenge. Unless, of course, you make use of digital twins.
Tangri, as well as Balaji Raghothaman, the chief technologist for 6G at Keysight, stated that digital twins were vital for testing out how well AI integrates into enterprise client’s frameworks.
“With all of these, the digital twin will be the centrepiece. It will be cloud native from day one. It’s a software solution, and it will be part of the whole offering that enables cloud,” Tangri said.
In March this year, Keysight unveiled its AI Data Centre Test Platform, allowing for the optimisation and validation of AI/ML networks that have been integrated into any organisation. The backbone for this, Tangri said, was the digital twin.
“A lot of industries already have a certain element of AI that they have been able to incorporate. The communication industry is playing a little bit of catch up on that because it’s more tradition bound.
“There is obviously a lot of scepticism and there is a vast mindset change that needs to happen. That’s going to happen only with confidence building, but also virtual testing with the digital twin. We think that the digital twin is an important answer to a lot of these questions,” Raghothaman said.
With this scepticism, the digital twin allows for better integration, as companies are more wont to integrate AI when they have verifiable proof that the integration would be seamless and improve the overall functioning of the company.
How Well Are Digital Twins Faring in the Testing Industry?
Well, with NVIDIA’s Cloud Platform, it’s safe to say that digital twins have become one of the better ways to test AI integration.
“This is why we are taking all our history of emulations, simulation solutions and creating very large complex digital networks, where our customers can actually experiment with their AI and see the outcome. This will help them gain confidence before they try to recreate it in a real data centre,” Tangri said.
Several companies have already begun using the tech indigenously to test out their own systems. Similarly, they have major competition with Ansys, Altair, and NVIDIA, who offer similar digital twin technologies for testing.
However, Keysight seems to be ahead of the curve in offering digital twin technology to test out AI integration within the communications sector, taking up a major market share in India. Additionally, their clients also range from Samsung Semiconductor India Research (SSIR) to Boeing and even NVIDIA itself, in offering the technology for testing semiconductors, vehicles and other products ahead of the launch.
Keysight may not be alone for long. The Digital Twin Market Report 2023-2027 found that the market for digital twin software grew up to 71% between 2020 and 2022. Additionally, as many as 63% of manufacturers globally are planning to develop or are already developing digital twins.
Additionally, another report by the BSI Group found that Indian businesses are more than ready to integrate AI into their operations, outpacing Chinese and American companies. The report, which was released on July 29 found that India had the most AI mature market, with metrics looking at AI investment, training, and communications, both internal and external.
Taking these into consideration, testing companies are likely to follow Keysight’s lead in leveraging their testing solutions and integrating digital twin technology.
The post Digital Twins Become the Centrepiece for AI Integration Testing appeared first on AIM.