Matt Jackson Of ‘Insight’ On The Future of Robotics Over the Next Few Years

An Interview With David Leichner

… Be Curious: In the robotics industry — as well as technology more broadly — you need to be curious and have an interest in uncovering solutions, especially for complex problems. Innovation is contingent upon an eagerness to learn and be curious. In technology, you can never stop learning.

With the shortage of labor, companies are now looking at how robots can replace some of the lost labor force. The truth is that this is not really a novel idea, as companies like Amazon have been using robots for a while now. What can we expect to see in the robotics industry over the next few years? How will robots be used? What kinds of robots are being produced? To what extent can robots help address the shortage of labor? Which jobs can robots replace, and which jobs need humans? In our series called “The Future Of Robotics Over The Next Few Years” we are talking to leaders of Robotics companies, AI companies, and Hi-Tech Manufacturing companies who can address these questions and share insights from their experience. As a part of this series, I had the pleasure of interviewing Matt Jackson.

Matt is a relentless problem-solver who is passionate about delivering solutions that help clients build operational excellence. He sees challenges as opportunities to push the boundaries and pave the way for technological innovation. As Insight’s VP & National General Manager, Digital Innovation, Matt builds on extensive consulting experience to help organizations accelerate their digital journey to modern business by tapping into Insight’s deep technical expertise, cloud and edge-based transformation solutions, and IT managed services.

Previously, Matt was the vice president and general manager of BlueMetal Architects when it was acquired by Insight in 2015. He was responsible for overseeing delivery and ensuring that their consultants had everything they needed to successfully deliver solutions that exactly meet their clients’ business and technical needs. Prior to BlueMetal, Matt worked at Burntsand/Open Text and Primix Solutions. He graduated with a degree in electrical engineering from Cornell University.

Thank you so much for joining us in this interview series! Before we dive in, our readers would love to “get to know you” a bit better. Can you share the most interesting story that happened to you since you began your career?

Early in my career, I was working at a robotics company building household products on the consumer side. The company also had a defense business, and one day as I was leaving work, I was approached at the exit by one of those massive army robots.

I froze when it walked up to me. It looked menacing — it was a military robot, after all — and I don’t know if somebody was controlling it or if it was AI. I wasn’t sure what to do, but funnily enough, the robot just greeted me and opened the door for me like that Boston Dynamics dog that went viral. A true gentleman! But I quickly walked to my car and drove away as fast as I could.

That was an early lesson that robotics can be fun, but it also can be scary in the wrong hands.

Can you please give us your favorite “Life Lesson Quote”? Can you share how that was relevant to you in your life?

It’s more of a mantra I live by: Always accept the challenge.

Every time I’ve had an opportunity to do something new — even if it’s daunting at first — my brain just does not compute the word “no.” Leaning into challenges has sometimes gotten me into some tricky situations, but it’s been the primary driver of my professional success.

Can you tell our readers about the most interesting projects you are working on now?

We’re working on a drone that can fly through aisles of a warehouse by using indoor mapping to understand where it needs to take inventory and what routes it should take to maneuver safely. The drone can count the number of products inside the warehouse, as opposed to someone manually doing an activity that’s monotonous, time-consuming and ultimately costly.

Another similar project we’re working on involves counting pallets using computer vision (my colleague Ken Seier’s team also built the AI model for this). Instead of having a drone fly around the warehouse, workers can use a tablet to quickly snap photos or a robot can move through the aisles on the ground to count products by grabbing images.

We’ve applied these computer vision solutions for uses in other industries beyond manufacturing. Using drones to help a national railway company automatically inspect track conditions, often in locations that are too dangerous for inspectors, and a renewable-energy utility to inspect turbines across wind farms. A large trucking company applies the technology using phones to quickly gauge tire tread wear on all their semis. These aren’t use cases that are romanticized (or demonized) in the movies; they are simply real-world improvements with significant ROI.

How do you think this might change the world?

Robotics isn’t just about the stereotypical “Robocop” humanoid, it’s about solving very real, table-stakes issues. In this case, cycle counts in the logistics industry are a big concern, taking a lot of manpower and time. Our solutions demonstrate how highly intelligent technology sits at the intersection of cognitive services — true AI and robotics — and everyday practicality. We can solve for highly repetitive problems that aren’t inherently complex yet really are the most vexing issues in business today.

These projects illustrate how automated technologies — drones, cameras, sensors, robotics — can perform repetitive or manual tasks, so that people are available for more meaningful work. This isn’t about eliminating jobs but allowing us to focus on higher level — and oftentimes safer — tasks.

Keeping “Black Mirror” in mind, can you see any potential drawbacks about this technology that people should think more deeply about?

The software that’s directing and powering these robots must be top of mind, especially when thinking about limiting bias and keeping data secure.

It’s important to remember intelligent tools are built on data collected from a variety of sources. It’s critical to understand where the data comes from, and what biases are included in each data sample to ensure autonomous systems — at the edge or at the core — are fair, ethical and don’t cause harm. However, any AI algorithm has the potential to make a mistake, even if it’s 99% correct (which is still usually more correct than a human).

You can think about this playing out at the doctor’s office during a check-up. While assessing a patient, the doctor takes notes during the appointment and could inadvertently be projecting biases — like gender bias — into their assessment. Then, those notes are harvested for diagnostic capabilities and could contribute to social inequities in access to healthcare or assistance down the line.

Any pattern of bias that exists in underlying data will be reinforced by AI, and any AI algorithm has the chance to make a mistake. Human intervention and continual assessment is critical.

What are the 5 things that most excite you about the robotics industry? Why?

1. Advancement of Technology: For someone like me, who is passionate about technology in and of itself, this is an exciting time for robotics from an innovation standpoint. There is spectacular work being done across the technology sector that helps advance critical areas for business, like digital transformation and IT modernization.
2. Freeing People Up: As I’ve mentioned before, robots can help people with automated, manual tasks that are often quite laborious for humans to complete. Having AI-backed robots support manual, otherwise time-consuming projects allows people to focus on higher level — and oftentimes safer — work.
3. Eliminating Bias: Although bias can still appear in these algorithms, robots are often less biased than people are — especially if the algorithms are properly designed and continually assessed, with eliminating bias being top of mind.
4. Optimizations and Precision: The optimization accompanied by robotics and automation often allows for finer, more precise, and accurate decision-making. We’ve seen it make a huge difference to our clients in the projects I mentioned above. In the warehouse counting case I mentioned, our client has actually used the automated, computer vision-powered inventory counting to enable their employees with disabilities to navigate their warehouses more safely and perform all the same tasks as their co-workers with highly accurate product counts.
5. Advances in Sustainability: Because of the ability for finer decision-making, robotics and autonomous systems have the potential to dramatically improve sustainability outcomes and help our planet. This can include limiting water use, supporting crop harvests, reducing factory waste, or optimizing electrical usage. For something like the Great Pacific Garbage Patch, I would predict that at some point robotics will get involved and help address the repetitive nature of the identification and collection process to help achieve its goal of cleaning up the ocean.

What are the 5 things that concern you about the robotics industry? Why?

1. Security Concerns/Hacking: As robotics and AI become part of our everyday lives and more pervasive across industries, we must anticipate and protect against security issues, especially regarding data collection and hackers. Hackers can target algorithms, feed different questions, uncover the data that was used to initially train it, and manipulate it for a different outcome that is not indicative of what it was supposed to represent.
2. Ethics & Bias: There can be ethical concerns that occur with robots, like any other technology or human decision making, especially as it relates to bias. Because of this, using AI in ethical, secure ways is a priority for any business. Any pattern of bias that exists in underlying data can be reinforced by AI, and any AI algorithm has the chance to make a mistake. Human judgement is a critical part of any equation.
3. Potential Impact on Job Creation: Although this is not a major concern given the kinds of jobs machines are often tasked with — think counting pallets or sending a real-time alert when a grocery store shelf needs restocking — there will need to be adjustments within the labor market to ensure there is minimal job loss.
4. Potential for Issues with Human Risk, Safety: Robots must be properly trained to ensure humans are not injured or physically harmed. For instance, imagine there is an industrial robot or a safety bot that is improperly trained to recognize when a human intervenes. It could lead to real risk. Education is involved, too, because there can be a tendency toward overreliance on technology. An extreme example of this is people driving autonomous vehicles putting too much trust in autopilot; it’s probably not wise to sleep behind the wheel while your car drives you to work. You still need to be aware of what’s happening around you and how the AI is responding to it.
5. Insufficient Training Data for Real-World Situations: Similarly, insufficient training data is also a concern, particularly for the first generations of autonomous bots that could encounter things they don’t expect or are not prepared for. For something like autonomous cars, if these technologies are not properly programmed to recognize the configuration of deviations from normal driving conditions, such as a car carrier loaded with cars “floating” past it, tunnels, or even windshield wipers, these vehicles will not react properly to whatever elements they encounter. There can be real safety or accuracy implications if the preparation of these technologies isn’t fully tested.

As you know, there is an ongoing debate between prominent scientists (personified as a debate between Elon Musk and Mark Zuckerberg), about whether advanced AI has the potential to pose a danger to humanity in the future. What is your position about this?

Any sufficiently powerful technology has the potential to pose a threat to humanity. AI is no different. The debate here stems from a Terminator-style, hostile Artificial General Intelligence (AGI). AGI is several years away and will require whole new fields of mathematics to be developed before it arrives. When it does it, we will have to come to grips with that new capability like we have for fire, agriculture, gunpowder, industrialization, nuclear fission and the Information Age.

What’s far more relevant is how bad actors are using AI today to orchestrate cybersecurity and social media attacks. The latter can sway public opinion to one extreme or another with devastating effects on regional and global economies. Also, the externalities of biased models — both models built on biased data and models constructed to benefit corporations at the expense of populations — can perpetuate inequality and discrimination with cold, inhuman efficiency. It’s in this sense that AI can impact humanity on a large scale in the short term while benefitting only a small fraction of the population.

In today’s environment, hackers break into the software running the robotics, for ransomware, to damage brands or for other malicious purposes. Based on your experience, what should manufacturing companies do to uncover vulnerabilities in the development process to safeguard their robotics?

For manufacturing companies to safeguard their robotics, it’s important to understand the network traffic. That way, anytime someone sends a command, it’s repetitive and predictable and can help monitor for any anomalies (think, “Oh, I know the process is supposed to do XYZ right now as opposed to doing ABC.”) This tactic can help identify any malicious hacking attempts, observe the robotics activities, prevent issues and shortcut the damages.

This requires organizations to implement best practices in networking and administration. Most attempted hacks are through remote infrastructures that are not maintained and/or are several patches behind. Hackers can find holes in the system and manipulate vulnerabilities. Because of this, having mature net-ops, upgrades and patches can be highly effective.

There must be a layered approach to security, in which the channel as well as the data, hardware, software and physical security are all maintained and protected.

Given the cost and resources that it takes to develop robotics, how do you safeguard your intellectual property during development as well as once the robot is deployed in industry?

Holistically, security is a layered approach, starting at the physical layer, and should be weaved into every aspect of a solution: OS, application/code, communication within components of the machine, as well as outside communication. The process of protecting IP depends on whether it is observable, written in code, visible in a physical form, etc. Using patents can provide one form of protection.

But today’s reality makes it extremely difficult to protect IP long-term. While you may be able to protect an algorithm if it lives in a secure core (on-premises or cloud), deployed robotics always can be reverse-engineered by a motivated third party. Physical structures can be dismantled and reproduced. Edge-deployed code can be interrogated and decompiled. Data and code can be encrypted at rest but must be decrypted at run-time and can be physically extracted from hardware.

The good news is that it takes considerable time and effort to do this, and only if your product is wildly successful. It took China 20 years to reverse-engineer the Israeli Harpy drone. Commercial-grade electronics are even more accessible.

What are your “5 Things You Need To Create A Highly Successful Career In The Robotics Industry”?

1. Be Passionate: Like any other industry, anyone considering a career in robotics must be passionate. It fuels the need to invest the time and energy into robotics and, ultimately, become successful. This passion isn’t just reserved for the outcomes of your hard work but the art of the process. I see this passion in my son, who is on a robotics team and absolutely loves it (we even need to take the robot away for him to go to bed!) He’s already so interested in how to leverage technology to create, innovate, build and solve. I hope that hunger stays with him in everything he goes on to do.
2. Be Curious: In the robotics industry — as well as technology more broadly — you need to be curious and have an interest in uncovering solutions, especially for complex problems. Innovation is contingent upon an eagerness to learn and be curious. In technology, you can never stop learning.
3. Invest Time, Energy: Robotics are complex systems, in terms of mechanical engineering, electrical engineering, computer science, etc. In order to build advanced systems across disciplines (blending the physical and the digital), you need to invest time in learning and understanding where you need to develop depth and hone the necessary skillsets. Robotics as a profession isn’t merely a hobby or a side job but requires a far deeper understanding of how everything works and integrates together.
4. Don’t Just Follow the Shiny Object: While the “humanoid” robot that can do backflips is cool and exciting, oftentimes the more real-world — perhaps mundane — use cases are the most fascinating and successful. That robot performing an inventory count can have an impressive ROI. Additionally, the machine learning models and cognitive services running these kinds of “routine” robots are extremely complicated. They can also drive the most valuable outcomes for society that move the needle and generate change.
5. Make Time to Disconnect: Whether you’re in robotics or any other profession — particularly in the technology sector — it’s important to be able to disconnect and explore nature and your surroundings. Technologists can oftentimes be too immersed and lose sight of what’s around them and how the technology they’re working on can impact the world around them. Without making the time for this, that’s when you start creating systems that will drive harm.

As you know, there are not that many women in this industry. Can you advise what is needed to engage more women in the robotics industry?

We’re seeing more women getting involved in robotics, AI and data, which is gratifying, but we have a long way to go. One big issue is that micro-biases and stigmatism persists starting in elementary school on up. Expectations that boys are good at math and science and girls are good at writing still exist. There is a lot of work to address STEM at the lower education level, but it will take time before those individuals enter the workforce. So, focusing mentoring efforts on those who identify as female coming out of college or who are looking to change careers can have a more immediate influence.

We just started a Diversity & Inclusion Group Mentoring Program at Insight that matches cohorts of mentees with professionals from different cultural backgrounds and organizational roles. It’s one way to invite new perspectives and expose our teammates to different career opportunities they may not have considered before. These are the types of door openers we all need to see more.

How can our readers further follow your work online?

Our team captures the innovative solutions we work on with our clients in the Insight Tech Journal and our own Medium channel called Insight From the Edge. We share our own perspective and lessons learned as we help organizations of all kinds figure out not just the IT but the right business strategy that accelerates their digital journey.

Thank you so much for the time you spent doing this interview. This was very inspirational, and we wish you continued success.