At least 80% of cybersecurity failures involve direct attacks on users' passwords, the World Economic Forum warns. This is a problem that's not going away any time soon. For Microsoft, as it explains in an official blog post, the solution may lie in a passwordless future, and it's hoping to ramp it up next year.

Whether it's storing personal information about addresses, finances, and highly private records, or running companies entirely virtually, the internet is where people live a good deal of their lives. And these people rely on passwords to guard their data and information against malicious actors. In response to hackers, some experts encourage the use of two-factor authentication, but that still isn't airtight.

"Passwords are a hassle to use," it notes, "and they present security risks for users and organizations of all sizes, with an average of one in every 250 corporate accounts compromised each month."

Security keys. For years now, Microsoft has been nudging people to adapt to a passwordless virtual landscape. To dispense with passwords without compromising security, the company offers physical encryption devices like the FIDO2 security key to open Hybrid Azure Active Directory Windows 10 devices.

Biometrics. Apart from security keys, Microsoft also emphasizes the need to try other techniques, like using biometric data to open devices as they do with mobile phones: fingerprints, iris scans, or Apple-style Face ID systems.

Biometric security measures come with their share of privacy concerns, though. But Microsoft is nonetheless eager to encourage a transition to biometrics and FIDO2 security keys. Because those are still more difficult to manipulate and compromise and less susceptible to the social engineering that so often sees passwords come undone.

The company wants this passwordless paradigm to be able to transcend different environments. Users should be able to, for instance, gain access to laptops and cloud-based apps with the same information that would let them access company buildings. Of course, this sort of cross-device and environment security makes ensuring it can't be compromised all the more critical.

Microsoft says its research has found that people are very open to the idea of ditching passwords for good. "Passwordless usage in Azure Active Directory is up by more than 50% for Windows Hello for Business, passwordless phone sign-in with Microsoft Authenticator, and FIDO2 security keys," according to the blog post.
​
But if this passwordless strategy is to become common and reliable in everyday use, Microsoft will have to go beyond its own users and bring rivals like Google and Apple on board, too. That could still take a little convincing.

Many Americans are very frustrated that they have to pay a mandatory router rental charge every month as part of their ISP bill. It gets added even if you purchase and use your own router in many cases. But as of Dec. 20, the practice is now illegal, and you can demand your ISP stop the charge.

As The Verge reports, your ISP can still charge for a router sent to a customer, but that customer should be able to return the router, have the fee canceled, and supply their own router instead. The charge is typically $10 a month, so the cost of buying a router is soon recouped. A good budget router can cost less than $50, meaning you'll start seeing the benefit of money saved in less than six months.
​
Don't expect ISPs to take the initiative here and cancel the fee automatically. If you are already using your own equipment, contact your ISP and demand that the fee is stopped. Send them a link to the bill if they protest. Alternatively, if you are using the equipment supplied by your ISP, do some research, see if there's a good alternative available and check the process for returning the ISP-provided equipment before attempting to cancel the fee. You want to avoid any internet downtime, especially as we are all so reliant on the internet for communication right now.

In the near future, your phone may take its 5G signal from the sky instead of a nearby mast on the ground. It's an innovative way to solve the problem of increasing connectivity without relying on thousands of terrestrial cell towers. The concept is known as a High-Altitude Platform Station (HAPS), and it essentially takes the cell tower from the ground and puts it in the sky.

The latest HAPS project to be unveiled is from Great Britain – Stratospheric Platforms and Cambridge Consultants. Recently, the pair revealed the core of its efforts, a special antenna and crewless aircraft, which it has been working on confidentially for the last four years.

How will it work? Instead of talking to a nearby tower to get its signal, your phone talks to a three-square-meter antenna attached to hydrogen-powered aircraft with a 60-meter wingspan, flying at an altitude of 12 miles (20 kilometers). The aircraft, or HAP, is expected to stay aloft for at least a week, all the while providing 4G LTE and 5G network coverage over an area of about 86 miles (140 kilometers).

Phones don't give off a powerful signal, hence the antenna array's size on the aircraft, which has 4,000 radios working together inside it. The processing power is similarly immense, helping to steer and direct the beams toward the ground even when the aircraft is shifting around. The cooling system has to work at high altitudes, minimize drag, and keep the weight manageable. It's an exciting piece of technology in itself.

The aircraft itself, which is made of a composite material, has already been certified for use, so it's deemed safe and ready for flight in civil airspace. There's no pilot, and ground-based operators will only be involved during takeoff and landing. The hydrogen power cell is not only environmentally friendly, as it only produces water vapor and emits very little noise, but it also gives off a lot of power — it has been tested to 50KW in labs already — for a long duration. That's a lot more than the low-power solar power systems used on other HAP vehicles.

HAPS systems are also expected to be cheaper to implement. Building each aircraft reportedly saves 70% over the costs of building and installing a traditional mast. Then there's the space-saving and logistical benefits. According to research quoted by the team, it's estimated that an impractical sum of 400,000 5G masts will be needed to cover the U.K., for example, and each aircraft could replace around 200 of those masts.

What can you expect from the connection? The aircraft should return a smooth Sub-6 5G beamed signal offering speeds of 100Mbits per second and 1m/s latency to devices connected to it, which don't need any special software or hardware modifications. While a fleet of HAPS aircraft could provide enough coverage for an entire country, what's very interesting is that the unique modular design of the antenna enables targeted coverage.

This means a signal from a part of the antenna could be focused on an individual area like a motorway or even a single vehicle driving on it. The team said a fleet of 60 aircraft could cover the U.K. in its entirety, and the idea is for networks to partner with dedicated "airlines" that operate the planes.

Potential problems. This new technology forces networks to rethink how they currently work. Network operators are already installing a 5G network using traditional masts. Convincing them to bypass this and adopt a sky-bound service will take a lot of work. However, the team calls its stratospheric network complementary to a traditional terrestrial network and says the costs involved will be temptingly less than creating a widespread 5G network using only ground-based masts.
​
When will it arrive? When can we expect to see Stratospheric Platforms HAPS system in operation? Deutsche Telekom is the project's initial investor and launch partner, and it has already tested an early version to show the system works. Stratospheric Platforms' aircraft is in preproduction now, with plans for the first prototype flight in 2022, while the antenna exists as a proof of concept and works with 3G, 4G, and 5G. It's anticipated the service will go live in 2024.

Remember when Amazon (quietly) announced its expansion to Sidewalk, back in September? Well, the feature is live for some in a new update for the Amazon Alexa app, and you might want to go turn it off. We covered it in Issue 7-25.

Sidewalk is a feature that extends the network coverage of your devices, particularly Ring surveillance tech (like its cameras, smart lights, and pet trackers) and Echo smart speakers. But it'll also share a small chunk of that internet bandwidth to provide the same services to your neighbors – so your privately-owned devices won't be so private anymore. 

Sidewalk has been slowly rolling out to Echo and Ring owners in the U.S. as of Thanksgiving, which users were made aware of via an email from Amazon. While the feature isn't up and running yet, the email essentially notifies users that it's "coming soon." But it's also the company's discreet way of letting you know the feature has officially been turned on. 

Amazon makes it easy to opt out if you're only just unboxing your shiny, new Sidewalk-compatible device. During the setup process, users are asked if they want to join the network via the Amazon Alexa app. However, if you already own one of the 20 Sidewalk-enabled products, it'll automatically opt-in for you. 
To disable Sidewalk, all you need to do is: 

• Update the Amazon Alexa app or double-check that you're on the latest version
• Open the Amazon Alexa app and tap on the More tab
• Then, tap Settings > Account Settings > Amazon Sidewalk and toggle off the Enabled button
  ​
  

Thankfully, the additional Community Finding feature – which "can help your neighbors find pets and important items connected to Sidewalk by sharing the approximate location of [your] device and other Sidewalk bridges you own" is disabled automatically.

Of course, if you'd like to use Sidewalk on either your Echo smart speaker or Ring security device, you'll be happy to know you're already all set for when Amazon officially launches the new feature.

One of my ‘go-to’ technology research firms is Forrester Research. Recently, the firm said 2021 would be the year that edge computing graduates from experiment to practically applicable technology, primarily driven by AI and 5G. Edge computing is a technology that places the processing close to the data source rather than concentrated processing at a distant central location.

Forrester has released a bundle of tech predictions for 2021, and part of it is a firm claim about edge computing: 2021 is the year it will finally become a real value. 

“Until now, edge computing was promising but still developing. In 2021, new business models will emerge that facilitate the deployment of edge in production,” Forrester said in a summary of its predictions. 

The new business models that will push edge computing “from science project to real value” in 2021 are based mainly on two factors. Forrester said: Cloud platforms having to compete with artificial intelligence and the widespread proliferation of 5G will make edge use cases more practical.

With those two drivers in mind, Forrester made five predictions about how the tech world will evolve in 2021 that will directly impact edge computing.

Edge Hosting Will Evolve into a Full-Fledged Marketplace. Content delivery networks (CDNs) like Akamai and Fastly, Forrester said, are starting to target edge computing demands, leading to them reaching out to colocation companies to find small, widely distributed data centers where applications can be hosted closer to the populations they serve. 

Even large colocation firms lack the localized presence needed to meet edge computing demands, Forrester said. Those same large companies often obscure the small local players required to complete their edge services.

“In 2021, colocation marketplace aggregators like Edgevana and Inflect will emerge as attractive options for the CDNs and global colocation leaders serving enterprise needs, even in rural locales,” Forrester predicts.

Kubernetes Will Dominate, but There Won’t Be an Orchestration Winner. Forrester predicts that lightweight Kubernetes deployments will end up accounting for 20% of edge orchestration in 2021, but that doesn’t mean the battle for edge orchestration will end. Canonical, Huawei, OpenStack, Rancher, and other companies are also trying to expand their lightweight, edge-optimized platforms, and competition will be fierce in the year to come.

AI Will Leave the Data Center for the Edge. Forrester predicts that the use of AI in edge computing will undergo a significant shift in 2021: Instead of machine learning models being trained in the data center, learning will start to happen at the edge. 

That shift will be possible thanks to new chips from Intel and Nvidia and new machine learning techniques like reinforcement and federated learning. “Edge application intelligence will blossom in 2021 to accelerate digital transformation, especially in industries that must bridge the physical and digital worlds in real-time,” Forrester said.

Private 5G Networks Will Spread. Nationwide 5G from major telecom networks won’t be sufficient to meet edge computing needs, Forrester said. In its place will be private 5G networks deployed by companies and developed by manufacturers like Ericsson, Huawei, and Nokia. 

In 2021, Forrester predicts, private 5G networks will be used in cases like factory floor automation, AR/VR for remote inspection, surveillance, quality assurance, remote monitoring, predictive maintenance, and employee safety.

Public Cloud Growth Will Slow While Edge Spending Will Grow. Forrester predicts the public cloud market will experience a growth decline from 42% in 2018 to 24% in 2022 due to market maturation. In its place will be an explosion of growth in edge computing, meaning more growth for companies that have invested in cloud-like solutions for edge computing and content delivery, not centralized data centers.
​
Public cloud entities won’t disappear, but they won’t dominate the future of distributed computing, Forrester predicts. “Their culture is based on massive data centers and tight control of the architecture; the exact opposite of what firms need to serve customers locally. Vendors with a winning edge strategy will do better.” 

Amazon is now starting to roll out its Amazon Sidewalk local networking system to customers in the U.S. Here's what you should know about it.

First announced in September 2019, Sidewalk is a new idea and long-term effort for extending the smart home to entire smart neighborhoods by using various local networking options. 

On November 24th, Amazon began sending out emails to Echo device owners letting them know that an update with Amazon Sidewalk will soon be rolling out. Though the system will take time to build out, the first steps are happening now.

With that in mind, here are some of the essential points that users should know about Amazon Sidewalk.

What is Amazon Sidewalk? Amazon Sidewalk is a new type of wireless network that makes smart home capabilities much longer ranged. A Sidewalk Bridge connects to your Wi-Fi network and essentially extends the connectivity range beyond what your router can output. In some cases, Amazon says this range could be half a mile.

It works by using various communication protocols, such as 900MHz radio signals and Bluetooth Low-Energy, for inter-device communications. The system will intelligently switch between these protocols depending on the range and power needed.

These Sidewalk networks work a bit differently than your home Wi-Fi, however. The bandwidth in a Sidewalk network is open for not just your own devices, but your neighbors', too. It's a bit like a local mesh network, but across a neighborhood. 

Most Amazon device owners already have a Sidewalk Bridge in their homes. Recent Echo and Ring devices will soon receive over-the-air updates that will allow them to work as bridges.

Benefits of Amazon Sidewalk. As mentioned earlier, one of the first and primary goals of Amazon Sidewalk is to extend the range of your smart home gadgets. What that looks like in practical terms could differ depending on your smart home setup.

A Sidewalk network could, as an example, ensure that outdoor security cameras or lights have a working connection even if they're far from your Wi-Fi router. It could also mean faster connectivity if a device connects to a nearby Sidewalk Bridge instead of attempting to connect to a router farther away.

If you happen to drop a Tile device while walking around the neighborhood, it could still be within the range of the local Sidewalk network — and it'll be able to connect to the appropriate servers using a neighbor's Sidewalk bandwidth.

Sidewalk will also make the onboarding of devices much quicker and simpler. And Amazon envisions other uses, too, such as a pet safety service called Amazon Fetch that alerts users if their pet wanders outside of a preset perimeter.

Are there any downsides to Amazon Sidewalk? For one, Sidewalk isn't a replacement for a home Wi-Fi network. The bandwidth available on a Sidewalk network is pretty small — Amazon says the maximum bandwidth is just 80Kbps, with a cap at 500MB. As such, it's only useful for low-power devices like smart locks, security sensors, and Tile trackers.

There are, of course, security and privacy concerns, too. If your dog has a Sidewalk-connected tag on its collar, it means that you may be sending Amazon the location, duration, and frequency of all your dog walks. Amazon does have a white paper that explains some of its Sidewalk-related security policies.

Of course, there's also the question of it being a shared network. Although Amazon says it will encrypt all traffic sent through a Sidewalk network, users won't know who is on a specific network or how much traffic their neighbor might be sending over it.

All in all, users will only be able to exercise marginal control over their local Sidewalk networks. There isn't currently a way for users to figure out which Sidewalk Bridge their compatible devices are connected to.
Can I opt out of Sidewalk? The downsides and implications of Amazon Sidewalk wouldn't be as pressing if it weren't enabled by default. The system is opt-out instead of opt-in.

Once Amazon enables Sidewalk, users will see an information splash screen that explains what it is and what it does. There, they'll have the option to disable it — it'll be turned on by default.

Users will also be able to disable the local networking system through the Amazon Alexa app. The option is located in Settings > Account Settings > Amazon Sidewalk.
​
If users do disable Sidewalk, their devices won't connect to their local neighborhood networks. On the flip side, neighbors won't be able to use their Sidewalk bandwidth, either.

The electric vehicle (EV) market's demand for ultra-fast charging has never been higher, as automotive batteries aim to replicate the convenience of filling up a tank at a gas station in minutes.

According to the US Office of Energy Efficiency and Renewable Energy (EERE), more than 80% of EV charging takes place at the driver's residence via basic wall outlets (100V/120V) or dryer outlets (220/240V). This makes sense as residential charging is convenient and inexpensive.

Problems arise when the EV is driven far enough away from the residence to warrant a recharge to get back home or continue the journey. This scenario is why fast-charging stations are growing in demand. Fast charging aims to recharge EV batteries within a short period of time, similar to refueling conventional gasoline vehicles. Today's fast charging typically takes about 20 minutes to charge up to 80% capacity.

The growth of fast-charging stations will help address the common driver concern of the limited range of their EV. The increase of fast-charging stations will mean drivers can easily refuel and significantly extend their vehicles' driving range.

But fast charging comes at a price. Until recently, many EV battery systems were not compatible with the fastest charging system. Fortunately, the major EV car manufacturers are investing in high-speed chargers with capacities of 150kW or higher and even faster, ultra-fast recharging technology.

Companies are working to address the issues between fast charging systems and battery capacity and safety. For example, Peter Kelly-Detwiler, Principal at NorthBridge Energy Partners, LLC, has been tracking the developments from SK Innovations.

"Why carry a huge canteen (or battery) on your hike if you know you can fill quickly at a stream every two or three hours on your journey," asked Kelly-Detwiler?  "Of course, the charging infrastructure will need to be there, with liquid-cooled cables, etc., but I do think a portion of the industry will end up at 350-400 kW capability to support the Hummers or SK Innovations supported EVs."

The need for liquid-cooled energy cables hints at another challenge for fast-charging EV stations, namely, their adverse impact on the electrical power system. Common effects include harmonic contamination in the transmission lines and high current, adding to peak-hour consumption concerns. The high power and current ratings of the recharging facilities require supervised operation at specially designed stations.

Types of Rechargers. There are two types of recharging "fuels," namely, alternating current (AC) and direct current (DC). A standard cord set is provided with each new vehicle that typically enables 3.6kW of charging via a conventional wall socket. Using this approach, the impact on the electricity grid is limited as the charging rate is so low.

There are three major categories of chargers, based on the maximum amount of power the charger provides to the battery from the grid, according to the EERE):

• Level 1: Provides charging through a 120 V AC plug and does not require additional charging equipment.  This level can deliver 2 to 5 miles of range per hour of charging. Most often used in homes, but sometimes used at workplaces.
• Level 2: Provides charging through a 240 V DC (for residential) or 208 V (for commercial) plug and requires the installation of additional charging equipment. It can deliver 10 to 20 miles of range per hour of charging. Used in homes, workplaces, and public charging.
• DC Fast Charge: Provides charging through 480 V AC input and requires highly specialized, high-powered equipment and special equipment in the vehicle itself.  (Plug-in hybrid electric vehicles typically do not have fast charging capabilities.) It can deliver 60 to 80 miles of range in 20 minutes of charging. This service is used most often in public charging stations, especially along heavy traffic corridors.

Depending upon the type of EV battery, how depleted it is, and its capacity, the charging rate amongst these three categories can be anywhere from less than 30 minutes to 20 hours.

Another type of charging, currently being researched by the Department of Energy (DoE) and provided to specific vehicle models Qualcomm and others, is cord-free wireless battery charging. For example, Qualcomm's Halo tech is currently fitted to the Formula E Safety Car and Medical car. The main advantage of wireless charging is convenience and not necessarily a faster charging rate.

A common concern with EV owners is the effect of faster-charging systems on the vehicle's battery life. The issue is that the chemical processes involved in rapidly charging a battery can lead to greater thermal loads that may degrade the battery faster than regular charges. The result could be reduced battery life and a decrease in driving range.

According to Jim Francfort, project manager in INL's Advanced Transportation Group and quoted from an INL article, "The value of building infrastructure is a difficult thing to measure. To put DC fast-charging stations along interstate corridors could encourage people to take longer trips even if they don't stop to charge."
The fact that fast charging is available gives EV drivers the confidence they might otherwise lack. This new breed of drivers will know that fast charging is available, like gas stations for traditional vehicles, if they need it.

Microsoft is urging users to abandon telephone-based multi-factor authentication (MFA) solutions like one-time codes sent via Short Message Service (SMS) and voice calls and instead replace them with newer MFA technologies, like app-based authenticators and security keys.

The warning comes from Alex Weinert, Director of Identity Security at Microsoft. For the past year, Weinert has been advocating on Microsoft's behalf, urging users to embrace and enable MFA for their online accounts.

In a blog post last year, citing internal Microsoft statistics, Weinert said that users who enabled multi-factor authentication (MFA) ended up blocking around 99.9% of automated attacks against their Microsoft accounts.

But in a follow-up blog post, Weinert says that if users have to choose between multiple MFA solutions, they should stay away from telephone-based MFA.

The Microsoft exec cites several known security issues, not with MFA, but with today's state of the telephone networks.

Weinert says that both SMS and voice calls are transmitted in cleartext and can be easily intercepted by determined attackers, using techniques and tools like software-defined-radios, FEMTO cells, or SS7 intercept services.

SMS-based one-time codes are also phishable via open source and readily-available phishing tools like Modlishka, CredSniper, or Evilginx.

Further, phone network employees can be tricked into transferring phone numbers to a threat actor's SIM card – in attacks known as SIM swapping – allowing attackers to receive MFA one-time codes on behalf of their victims.

On top of these, phone networks are also exposed to changing regulations, downtimes, and performance issues, all of which impact the availability of the MFA mechanism overall, which, in turn, prevents users from authenticating on their account in moments of urgency.

SMS and voice calls are the least secure MFA method today.

All of these make SMS and call-based MFA "the least secure of the MFA methods available today," according to Weinert.

The Microsoft exec believes that this gap between SMS & voice-based MFA "will only widen" in the future.
As MFA adoption increases overall, with more users adopting MFA for their accounts, attackers will also become more interested in breaking MFA methods, with SMS and voice-based MFA naturally becoming their primary target due to its extensive adoption.

Weinert says that users should enable a more robust MFA mechanism for their accounts, if available, recommending Microsoft's Authenticator MFA app as a good starting point.

But if users want the best, they should go with hardware security keys, which Weinert ranked as the best MFA solution in a blog post he published last year.
​
This preference for app or security key alternatives for MFA shouldn't mean that users should disable SMS or voice-based MFA for their accounts without substituting another MFA approach. SMS MFA is still way better than no MFA at all.

Zoom calls will now come with the option of live captions, in a move that’s likely to make life easier for remote workers whose attention spans suffer steep declines during online meetings.

In March 2018, we published in issue 4-39 an article “Transcribe Your Meetings in Real Time” introducing you to Otter.ai, a Los Altos, CA start-up. Now the company is expanding its technology to enable speakers on a Zoom call to see their words turned into accurate captions in real-time. This should put an end to the unfortunate miscommunications frequently caused by remote collaboration tools.

So, there should be no more excuses for misreporting the numbers presented by your sales team or missing the list of targets put forward by your manager.

Captions will appear directly within the call, with a couple of seconds of lag. Presumably, they will be accurate enough for crucial information to come out in plain text consistently. 

The new feature will be incredibly helpful to users with accessibility needs and non-native English speakers struggling to make out the meaning of a sentence. Otter.ai currently only supports the English language but can handle a variety of accents, including southern American, Indian, British, Scottish, Chinese, and various European accents.

Two years ago, Otter.ai launched its popular speech-to-text software. Available as a mobile app or as a web-based tool, the technology soon started supporting online conferences, offering users the option to turn Zoom cloud recordings into written conversations to keep a record of their virtual meetings. 

Earlier this year, Otter.ai launched Live Notes – a new feature that enables users to open a live transcript of the call during a video conference in a separate shared file, which transcribes what is being said in real-time.  
Based on a sophisticated algorithm, Live Notes can separate human voices to identify different speakers and includes their name in the transcript to indicate that a given participant has started intervening. Users can then go back to the file to check a detail if they have missed a sentence or jumped late into the call.

Therefore, the new announcement builds on top of Live Notes, integrating the transcribed quotes directly into Zoom’s platform during a virtual meeting. In a demo call showcasing the technology, Otter.ai’s founder Sam Liang said, “Now, you will have Live Notes still going on in the background, but then you will also have the captions put down in the call. And there’s a pretty broad range of people that this will be helpful too.”

“It’s definitely a great help for people with a hearing disability, but also for international, distributed workforces who don’t speak English as their native language. And education as well: online classes could benefit from captions, on top of the Live Notes that they can go back to, to facilitate learning.”

The transcription is not precisely pitch-perfect: some sentences don’t make sense, and words occasionally come up deformed. Overall, however, Otter.ai’s algorithm, especially given the tool’s ease of use and accessibility, appears to be pretty accurate – an assessment echoed by most online reviews and user experiences.

Liang is confident that the technology’s accuracy is only improving as more users get on board, providing more training data for the speech-to-text algorithm and helping the AI work its way through background noise and strong accents.

By considering the context of an entire sentence, rather than working on a word-by-word basis, the AI can make more accurate decisions. 

Similar methods have sparked the interest of the industry’s most prominent players, with IBM now offering a cloud-based, highly accurate speech-to-text platform as part of Watson’s services. In contrast, Amazon Transcribe provides an API for automatic speech recognition.

However, Otter.ai is arguably the most consumer-facing technology out there. Liang confirmed that the company is now working on a smoother integration with platforms like Microsoft Teams, Google Meet, or Cisco Webex to open up access to the transcription and live-captions features.
​
In Zoom, live captions are available now for Otter customers paying for a Business plan, as well as for Zoom Pro customers.

Students will have to “learn how to learn” instead of focusing on skills and specific tools. Most content will become obsolete within two years, said Dr. Simon See, head of computing giant NVIDIA’s artificial intelligence (AI) technology center.

“Technology is moving so fast, and you have to keep adopting new skills. What you have to learn is the methodology of learning new things. You must learn how to learn new things,” he said.

Dr. See spoke in a panel discussion at the launch of Nanyang Technological University’s (NTU) Deep Learning Week (DLW) recently.

Minister-in-Charge of the Smart Nation Initiative Vivian Balakrishnan was the guest of honor.

In a video call to students, staff, alumni, and members of the public, Dr. Balakrishnan emphasized the need for university students to develop “deep technical skills” in areas such as data analytics, machine learning (ML), and artificial intelligence (AI).

“You are the dynamos that will drive our transformation and push the boundaries of how we can leverage technology to transform our economy and make a real difference,” he said.

This is the second edition of NTU’s DLW, organized by the Machine Learning and Data Analytics Lab at the NTU School of Electrical and Electronic Engineering (EEE).

During the DLW, there were AI-related workshops as well as a virtual AI career fair for students, ending with a machine learning hackathon themed “AI in business and economics,” said NTU in a statement.

Along with Dr. See, the other panelists included Mr. Sim Kai, deputy director of the national AI office, Dr. Laurence Liew, director for AI industry innovation at AI Singapore, Ms. Jane Shen, chief scientist and managing director at Pensees, Dr. Pan Yaozhang, head of data science at Shopee, and Dr. Yap Kim Hui, associate professor at NTU’s School of EEE.

Distinguished panel members included Ms. Jane Shen, Dr. Laurence Liew, Dr. Pan Yaozhang, NTU EEE chair Tan Yap Peng, Dr. Simon See, NTU EEE student committee president Duan Jiafei, Mr. Sim Kai, Dr. Yap Kim Hui, and moderator Dr. Wesley Tan.

Moderated by Dr. Wesley Tan, senior lecturer at NTU’s School of EEE, the panel discussed the Covid-19 pandemic’s impacts on the economy and digital transformation.

Panel members concurred that while Covid-19 had forced companies to speed up digital transformation, most trends were already in motion since last year.

The panel also believed that passion was more important than academic background when pursuing a career in AI.

Dr. Liew said that computer scientists were a minority among the applicants in his organization’s AI apprenticeship program, with social scientists and many from other disciplines who had also applied and were doing well.

“A lot of the most successful tech people in the world do not come from a computer science background,” he noted.
​
“It doesn’t really matter what your first degree is in. As long as you are passionate, you can get into AI.”