Real experts. Real ideas. Real conversations. Protolabs presents… Real Talk.
Welcome to Real Talk, a thought leadership audio series that delves deep into the transformative changes and challenges within the manufacturing industry.
This series of conversations explores how industries will confront and overcome the biggest questions facing them in 2024 and beyond. Three of the most influential voices within a given topic cut through the noise and deliver a debate rooted in practical insight, not speculation.
What is artificial innovation, and why will it transform business?
Artificial intelligence is rapidly being integrated into almost every area of our lives. And in the manufacturing industry, it’s already having a profound impact. The World Economic Forum predicts that the market for AI in manufacturing is expected to grow to $20.8 billion by 2028.[1]
So, how can the manufacturing industry use AI as an enabler for innovation rather than as a disruptor of growth?
In this audio feature, three global senior thought leaders discuss why manufacturers must go beyond AI implementation to practise artificial innovation.
[1] https://www.weforum.org/agenda/2024/01/how-we-can-unleash-the-power-of-ai-in-manufacturing/
The Speakers
Synopsis
Artificial intelligence is rapidly being integrated into almost every area of our lives. And in the manufacturing industry it’s already having a profound impact. The World Economic Forum predicts that the market for AI in manufacturing is expected to grow to $20.8bn by 2028.[1]
But in their rush to implement AI technologies, are manufacturers running before they can walk? Or are they successfully using AI to become more innovative?
In this audio feature, host Meg Wright, Head of Innovation at FT Longitude, asks why manufacturers must go beyond just implementing AI to practise artificial innovation.
Joining Meg are:
- Aric Dromi, former in-house futurologist for Volvo and founder of RETHNK.GROUP
- Umbar Shakir, Digital Lead and Partner at Gate One
- Carlos Sentís, Founder and CEO of Improve X AI
[1] https://www.weforum.org/agenda/2024/01/how-we-can-unleash-the-power-of-ai-in-manufacturing/
Transcript
VO: You’re listening to The Real Talk series of conversations from Protolabs, produced in partnership with FT Longitude.
Meg Wright: The AI revolution has arrived.
Artificial intelligence is rapidly being integrated into almost every area of our lives, from healthcare to finance, education to entertainment, and everything in between.
In manufacturing, it’s already having a profound impact.
And as AI’s capabilities continue to grow, manufacturers are racing to understand how they can use it to transform their efficiency, productivity and profitability.
According to the World Economic Forum, 68% of manufacturers across all sectors have implemented some form of AI in their operations.[1]
And the global AI manufacturing market? Well, that’s expected to grow to an estimated value of $20.8 billion USD by 2028.[2]
As companies rush to take advantage of this promise, a considered approach will be vital.
Umbar Shakir: I do actually find that some of my clients are trying to run before they can walk. So while AI presents us with amazing opportunities, not every business problem has an AI shaped answer.
Aric Dromi: You have so many organisations today that are fast jumping into this swamp of solutions without actually asking the relevant questions. And I think when we are shifting into a world that is dominated by algorithms, the questions are becoming so much more important right now.
Meg Wright: New research from Protolabs reveals that the promise of an AI-disrupted industry has yet to be fully realised.[3] But there’s no doubt that change is upon us.
Carlos Sentís: A lot of jobs are expected to be destroyed and a lot of sectors are going to be disrupted. And if we don't innovate fast enough, it's going to be very difficult to compete or to just keep our economies growing.
Meg Wright: So, how can the manufacturing industry seize the opportunities AI presents?
I’m Meg Wright, Head of Innovation at FT Longitude. In this audio feature I’ll be exploring why manufacturers must embrace artificial innovation and how it will help them to flourish in the years ahead.
[1] www3.weforum.org/docs/WEF_Harnessing_the_AI_Revolution_in_Industrial_Operations_2023.pdf
[2] https://www.weforum.org/agenda/2024/01/how-we-can-unleash-the-power-of-ai-in-manufacturing/
[3] https://protolabs.gcs-web.com/news-releases/news-release-details/protolabs-3d-printing-report-reveals-accelerated-industry-growth
Chapter 1: Why AI in manufacturing?
Meg Wright: Ask anyone in manufacturing—from automotive to life sciences—and they’ll tell you that AI is integral to the future of the industry.
Carlos Sentís: Well, the first thing that we need to know is that artificial intelligence has now the ability not only to predict or to find patterns, but also to generate with generative AI to generate any type of modality. So text, numbers, audio, video, 3D, code.
Meg Wright: Meet Carlos Sentís, founder and CEO of Improve X AI, an executive upskilling and strategic advisory firm focused on the practical applications and risks of artificial intelligence.
Carlos Sentís: How does that affect manufacturing companies or any company at all? Well, first of all, if you have products, then you can design more products, you can iterate on the products that you design. You can iterate on the design of the packaging, but then you can also iterate on the machines that you use. You can improve the machines that you use. You can reduce the cost of both the efficiency of your own factories or the whole supply chain. And so both from the generation perspective, the ability to create more and better and the savings part, the optimization, the better resource allocation that you can do is massive.
And so basically what artificial intelligence does is enables for all of the combinations to be explored and then deploy algorithms to perfect or to optimise and find what is the optimal solution for a problem or what is the best candidate that we have.
Meg Wright: These capabilities help manufacturers improve efficiency, productivity and production. And it’s what Umbar Shakir calls “everyday AI”.
As Digital Lead and Partner at digital and business transformation consultancy, Gate One, Umbar is well-versed in tech-enabled change.
Umbar Shakir: I head up our technology practises, so that's digital strategy, tech platforms, agile and product transformation and data and AI. But what I’m seeing is that most of my clients are focused on everyday AI. So they are looking at that productivity automation. And I've seen some really interesting examples of that productivity gain.
One is with a North American pharma company, what they've done is, and actually we helped them build these tools, they would have to submit marketing literature for a regulatory compliance review and sign-off. So you couldn't change anything on a website without it going through some sort of regulatory process. And that was with external bodies. So they've been doing this process for a number of years. They submit their content, they get an audit back. And by training an AI model in all of the data that was ever submitted, plus all of the decisions that were ever made, now what Generative AI can do is if you submit your content based on the learning that it's had from previous audits, it will give you a red, amber, green status on whether your content might pass a regulatory review and it's saving them millions.
Aric Dromi: I always say that the purpose of any company is the bottom line is to make money.
Meg Wright: Here’s Aric Dromi, strategist, creative advisor and founder of RETHNK.GROUP.
As the former in-house futurologist at Volvo, Aric knows all-too-well the challenges the automotive industry is up against. In particular: how to come out on top in the transition from combustion engines to electric vehicles.
Aric Dromi: And there are two types of ceilings that we need to cater for. There is a glass ceilings that it's easy to break, but there is also a concrete ceilings, that we break our head instead of the ceiling itself. And we are fast reaching a point which we cannot really introduce new efficiency models, we cannot improve on efficiency, so much more beyond what we are capable of doing today. So we need to shift discussions from value to impact. What is the new impact line that we need to create right now that will enable us to scale into a future?
Meg Wright: The answer: Artificial innovation. This is the lever that will elevate manufacturers from simply delivering efficiency to pioneering new ideas.
Chapter 2: What is artificial innovation?
Meg Wright: So, what exactly is artificial innovation? And how can it help manufacturers preserve and grow their business for years to come?
Carlos Sentís: Artificial innovation is leveraging artificial intelligence to expand our ability to innovate and find solutions to problems. Innovation in its essence is improving the state of things. It is finding new ways of doing things or just creating new opportunities to grow our businesses or also to tackle some of the challenges we face. And I think examples speak very loudly about this.
One example is the difference between what a human with his intuition can do to create new products or new services versus what you can do when you have the power of AI. And an example of that is the work done by DeepMind that is now part of Google led by Demis Hassabis with GNoME, which is material science innovation. And what they did was instead of just creating one new research about new potential materials, they did this kind of combinatorial where they put together all the different types of possibilities to try to find the best new materials that they could.
And in human history we have in total discovered around 200,000 materials out of which 50,000 are stable. In just one year, they have been able to discover 2 million new materials out of which 400,000 are actually stable or they have the potential to be stable.
Meg Wright: So artificial innovation is an intentional expansion of AI’s capabilities, taking it from the operational to the aspirational.
It is a partnership between humans and machines that aims to set and achieve bold ambitions.
Umbar Shakir: What's happening now with artificial innovation is with Generative AI, the machines are learning our language. So, natural language processing, they are learning from all of our vast amounts of digital data that we've been accumulating for the last 30 years, and they're now able to generate and synthesise new data back to us.
So there's a real marked difference in how the technology is developed, in what it learns, in how it's learning and how it presents back to humans. So we're fundamentally redefining the whole relationship that humans have with machines.
Meg Wright: Despite this enormous potential for change, some manufacturing companies are missing the mark. In Aric’s experience, their strategies are technology-centric rather than human-centric.
And, as Aric explains, this risks undermining not only the innovative potential that AI has to offer—but also the valuable opportunity to improve end user experience.
Aric Dromi: We live in this buzzword reality that is driving our decision making. We are driven by technology, not by the user experience. And in the automotive industry it's very, very much clear, and you have things that are so simple that can improve the journey dramatically, but they're not even there at the moment of time.
User experience, for me it's to know that I don't need to wait for parking, is that I can go to the city with my wife to have a date night and drink, and not worry, because we are going to have a concierge that is driving us home in our car. User experience is, I know that four times a year on my so-called subscriptions, I have someone that drops me in the airport and pick me up from the airport. User experience is about, "Hey, don't worry, you are travelling right now to Australia. We will come the day before and pick up your luggage, it'll meet you in the hotel. You don't need to stand." That is user experience. Mobility is not just when you step into the car, start the car and stand in traffic. It's so much more holistic than that. And the automotive companies, they should revolutionise this industry.
We don't even have sufficient enough business cases to justify investment in infrastructure, which enable autonomous driving cars. We don't even have that today. So before we rush to implement ChatGPT inside of our cars, we need to ask why. What the user is going to do? Where is the value that is going to be given to you by actually having this type of interaction in the vehicle? We don't have the right questions in this industry at the moment.
Meg Wright: Compounding this challenge, Umbar raises a further concern: that businesses may be rushing into AI transformation without paying attention to their ethical responsibilities.
Umbar Shakir: If you try to jump to trying to put in AI before your data or your processes are ready you’re going to amplify some of the red herrings in your business, you’re going to amplify some of the biases that are in the data. So I do caution, which is weird for a management consultant to go, “Let’s slow down,” but I call it the artificial innovation cha-cha-cha, like you have to do one step backwards to go two steps forward.
And I can't stress this hard enough, that if you want to be ethical, you have to be inclusive. You have to start to really look at what is in the large language model. And that's really hard to do because big tech are not disclosing their sources for large language models, which is why then the only way of understanding the level of bias is to test different large language models.
But we know already from the last 30 years of the digital data sets that we have, they're largely western data sets. We know that they are hypersexualised, particularly around imagery of women. We know that they are biassed and prejudiced against Black ethnic minority communities. And we know that there's a lot of vitriol and hate speech in a lot of that data. So when I said that we were learning language of the machines, now they're learning ours, we're not teaching it with the best datasets.
Meg Wright: If true innovation relies on diverse thinking, this must also be the case for artificial innovation. So how do we create responsible and innovative AI strategies? For Carlos it comes down to asking the right questions.
Carlos Sentís: Does your leadership understand the opportunity and the risks and is embarking everybody on that journey of learning about it and trying to see where can we use it and how should we use it? And what are the limitations that the current technology has? And what are the best uses for us? Because it's not necessary for us to do everything that is possible, but we really have to focus on innovating in what's really value adding and discarding all the other things. Because a lot of the problems that come with innovation is that you don't know exactly what to say no to.
Chapter 3: How can businesses seize this opportunity?
Meg Wright: So how can manufacturing businesses adopt a truly innovative mindset and seize the opportunities presented by AI? And what does this look like in reality?
Umbar Shakir: I've also spoke about human work in terms of craft and graft. And graft is the input type stuff that we do, research, analysis, collation. And craft is where we tap into whether it's our humanity and our emotions, our values and our moral compasses and then our brains and our creative solutions. And it's how we solve problems.
Meg Wright: To begin with, leaders must expand their mindset by continuously educating themselves on the potential of AI. New use cases are constantly emerging, and with them new opportunities to innovate.
Take, for example, embodied AI, which is now revolutionising the field of robotics, as Carlos explains:
Carlos Sentís: When you plug AI into a robot, you can do almost anything. So you can plug these systems both in drones but also in robots. And robotics are not just what people think about when it's thinking about just a manufacturing facility, there's also humanoid robots that are capable of doing more and more tasks that now have vision and understand their surroundings, that they can speak, they can move around.
But not only that, there's even new generations of robotics that allow for flexible robots. Robots that are able to fuse themselves, turn into liquid and move somewhere else and then become solid again. They have now the ability to expand their size and some of them, again, are flexible. So they're not the super rigid clunky robots that we thought of before, but rather we have an infinite possibility of robots with all these capacities of understanding the surroundings, of speaking, of gathering information. And all of that can be used in every part of the process of manufacturing.
Meg Wright: It is precisely these types of innovations that allow humans to take charge of the AI partnership, shifting the focus from improving efficiency to dealing with real-life pain points. Here’s Umbar again.
Umbar Shakir: Building a culture of artificial innovation where you are looking at the various types of AI and how they might be applied to your business needs some frameworks.
How do you discern the results that you are getting from different models? Because different models give you different results for different types of use cases. And so it's really important that you have that framework. And then fundamentally, what's the outcome you are trying to drive to? So an adaptive strategy framework is really useful here, because the rate of innovation is so fast that you don't want a static two to three, five year strategy anymore. You need to have something where you can pivot.
And so set your strategic objectives, have very clear outcomes when it comes to customer, commercial, capability and colleague. Prove the concept, incubate ideas, test and learn, but then be willing to either pivot on your strategic objectives or pivot on some of the outcomes you're trying to achieve.
Meg Wright: To ensure artificial innovation really is a strategic priority, Aric says business leaders must consider how much energy they are putting into day-to-day business needs and the bigger picture.
Aric Dromi: There is a model that I really like. It's called the 90-9-1. Spend 90% of your time catering to everything your company needs. That's fine. You can use a lot of technologies to automate these needs. Spend 9% of the time talking with your executives about what you actually want, and then spend 1% of your time focusing on what you wish for.
When you put 100% of the management power of your company up to the point of sale, you can never understand how tomorrow is going to look like. It's not going to happen. So you need to take a step back and say, "How can I use technology to automate my day-to-day needs, so I can have more time to think later on?"
For me, this is what artificial innovation is all about, is to be able to streamline, to be able to analyse large sets of data to capture the user feedback loop and identify patterns in it, so you can feed it back into the product, into the service in real time. It's a partnership between a creative mind and an engineer mind that is represented by an algorithm.
Conclusion
Meg Wright: In our increasingly digital and disruptive world, a company’s survival depends on its ability to innovate and to use AI intelligently. Now it’s up to manufacturers to embrace artificial innovation to create new opportunities for themselves and their people.
Carlos Sentís: We're going to need to be more proactive, and I hope that organisations, both public and private, and policymakers help our societies to learn this information and to be more proactive, to create new opportunities for themselves. Because if we are self-reliant, we're in a much better position. If not, I think that we have a huge risk of a painful transition that can lead to violent protests very soon.
Meg Wright: And while the manufacturing industry’s AI race is based on good intentions, leaders must take care to be guided by strategic insight and not get swept along by the hype.
Umbar Shakir: It's not the technology that worries me, it's people that worry me. It's business leaders who, if you are caught up in the hype, thinking that I can make lots of teams very lean and do this really quickly. And I worry about that because I think they're going to be very underwhelmed and they're going to realise that they've maybe even wasted money and resources on some of these things if they don't get that upfront strategy and planning right for how this technology can really help their people, help the quality of work, help solve the right business problems. But let's try not to shoehorn this in just for the sake of technology.
Aric Dromi: The automotive industry is very much driven by technology. Everyone in the automotive industry is saying, "We're a mobility company, we are a technology company, we are not a car company." But what's the purpose of a car in the end of the day? And what's the purpose of the value of the car? And if we start to ask these questions, it will help us redefine the entire business model landscape of the automotive industry.
And I don't see these questions happening right now and we have this bunch of executives that are rushing in to implement algorithms everywhere, not understanding the long-term consequences that that path is going to have.
It is based on content, not based on context. But humans are driven by context and intention, not by content. And I don't like this idea that content is the king, because it's not. It's the context that is the king. So if we do want to use AI in a much better way, it needs to be driven by context and intention.
Meg Wright: I’m Meg Wright. Thanks for listening.
VO: You’re listening to The Real Talk series of conversations from Protolabs, produced in partnership with FT Longitude.