One of the biggest takeaways from the Canadian Telecom Summit was that customer behaviours have changed significantly because of the pandemic — and there’s no going back.
We’re seeing huge increases in virtual interactions, including collaboration and video conferencing, as well as on-demand video — and this digital disruption is expected to continue post-pandemic. While some workers will return to the office, we’ll also see a shift from WFH to WFA (work from anywhere).
This is accelerating digital disruption: everything from re-engineered networks and public cloud adoption to resilient supply chains, AI and automation, and 5G for a WFA economy.
Responding to the new normal
Jump-starting the economy through disruptive technologies such as 5G, AI and blockchain was the subject of a virtual panel at the summit, hosted by Namir Anani, president and CEO of the Information and Communications Technology Council (ICTC).
“This discussion comes at a very critical time … specifically as businesses and companies are trying to grapple with this environment and respond to the market needs of this new online normal,” said Anani. There’s also recognition that post-pandemic recovery will be “innovation intensive.”
Disruptive technologies such as 5G, AI and blockchain are “very much linked to big-picture thinking and disruptive innovations” in everything from connected vehicles and connected health to smart cities and Industry 4.0.
“But here’s the catch,” said Anani. “Are we looking at these technologies and applications in the right way? Are there any alternative applications or alternative technical technologies that complement and add to the space in this environment?”
5G and digital disruption
Germany, like its counterparts in the European Union, is facing a problem: Because of the speed of technological development, regulators and innovators are distancing themselves from each other — even more so than they have in the past.
Dr. Thomas Könen, head of the Department for Digitalisation and Innovation with the Bundesverband der Deutschen Industrie (BDI) – Federal Association of German Industry, joined the panel from Berlin. Germany’s digital infrastructure is “very, very slow,” said Könen, and “we hope that 5G is going to be a leap-frogging step for us to jump back into the game.”
As part of the move toward Industry 4.0, Germany is providing 5G frequencies not only to telecommunication providers but also to industrial companies (over a local campus network, not a regional one). This broadband frequency band is reserved only for industry; to date, it’s been rolled out to 50 companies.
There’s also the issue of digital sovereignty. “It’s the question of who should be providers for software and especially hardware for 5G networks in the future to come,” said Könen.
5G to drive innovation
But while Germany took first place in the 2020 Bloomberg Innovation Index (for its prowess in value-added manufacturing, high-tech density and patent activity), the U.S. fell a notch from the previous year to No. 9 and Canada came in at No. 22.
“So it’s relatively mediocre middle-of-the-pack performance (for Canada) and we haven’t achieved yet our full potential as a country in terms of the innovation economy,” said Jean-Charles Fahmy, CEO and strategic advisor for CENGN, Canada’s Centre of Excellence in Next Generation Networks.
Fahmy said Canada needs to work on three pillars to improve those metrics. The first is ubiquitous connectivity to create a “data-gathering engine” — and 5G will be a key component of that, along with cloud computing, edge computing and software-defined networks.
The second pillar is AI and predictive analytics (to deliver value from big data generated by ubiquitous connectivity) and the third is a trust framework (securing data and infrastructure, as well as privacy laws, sovereignty rules and ethics).
“Those three forces together can be a significant engine of economic and social prosperity,” said Fahmy. “It’s how we are going to get the full value of digital transformation and connectivity for all citizens.”
The ethics of AI
When it comes to the transmission of data, “faster pipelines are going to make a huge difference,” said Cameron Schuler, chief commercialization officer and vice-president of industry innovation with the Vector Institute, an independent, not-for-profit corporation dedicated to research in the field of AI.
“From my perspective there’s a few technologies that I think are going to be incredibly valuable,” said Schuler. “One of them is going to be federated learning — so the ability to take smaller models and have that feed into large models, as well as being able to retain data privacy.”
That all ties into AI — and with that comes the issues of legal requirements and ethical considerations. Scott Zoldi, chief analytics officer at FICO, a Silicon Valley-based pioneer in the use of predictive analytics and data science, said we’re seeing “this trend of people retreating to safer technologies and less use of data until they get these ethical AI considerations resolved.”
Many organizations are feeling ill-equipped to consume this data at scale, he said, and monetize it or make predictions around it.
“It really talks in my view to the fact that AI … has to mature in the same way that software practices have very strong standards around agile development and code testing and regression testing,” said Zoldi. “We don’t generally see that in AI and as we consume more of this data and we rely on AI it’s going to be fundamentally important.”
Regulating AI with blockchain
Zoldi expects it will take at least five years — or maybe more — before we can do this responsibly (and demonstrate that to customers and regulators). It requires everyone to get on the same page, since different governments and businesses have different philosophies with respect to how to drive AI innovation in the context of regulation.
“In the U.S. there’s sort of a view that you just innovate and we’ll worry about the regulation later,” said Zoldi. “We need to look at AI the same way we would look at the development of software where we’d have … security by design,” adding that “this can’t be up to the data scientists’ artistry.”
The best way to move forward and ensure proper adoption, he said, is to construct an AI governance policy around development, testing and monitoring. FICO, for example, uses a model governance blockchain, which is “basically an enforcement vehicle for a corporate standard around AI development so that we understand the steps that the corporation deems necessary,” said Zoldi.
Using blockchain, they can ensure no steps are skipped in the development of their analytic models, and they can go back and refer to it later when the model is in production.
“We’re looking at tools like blockchain as one of those ways to demonstrate proof of work, where you can go back to this immutable chain of the history of events that occurred during the model development,” said Zoldi.
While 5G, AI and blockchain aren’t new to us, there’s still much work to be done from a regulatory, security and ethical standpoint. If one thing was made clear at this year’s Canadian Telecom Summit, it was that the pandemic is accelerating digital disruption — and the status quo is no longer an option.