AI, Automation and Industrial Innovations - April 2022

AI, Automation and Industrial Innovations – April 2022

How to investigate when a robot causes an accident – and why it’s important that we do

Robots are featuring more and more in our daily lives. They can be incredibly useful (bionic limbs, robotic lawnmowers, or robots which deliver meals to people in quarantine), or merely entertaining (robotic dogs, dancing toys, and acrobatic drones). Imagination is perhaps the only limit to what robots will be able to do in the future.

What happens, though, when robots don’t do what we want them to – or do it in a way that causes harm? For example, what happens if a bionic arm is involved in a driving accident?

Robot accidents are becoming a concern for two reasons. First, the increase in the number of robots will naturally see a rise in the number of accidents they’re involved in. Second, we’re getting better at building more complex robots. When a robot is more complex, it’s more difficult to understand why something went wrong.

Most robots run on various forms of artificial intelligence (AI). AIs are capable of making human-like decisions (though they may make objectively good or bad ones). These decisions can be any number of things, from identifying an object to interpreting speech.

AIs are trained to make these decisions for the robot based on information from vast datasets. The AIs are then tested for accuracy (how well they do what we want them to) before they’re set the task.

AIs can be designed in different ways. As an example, consider the robot vacuum. It could be designed so that whenever it bumps off a surface it redirects in a random direction. Conversely, it could be designed to map out its surroundings to find obstacles, cover all surface areas, and return to its charging base. While the first vacuum is taking in input from its sensors, the second is tracking that input into an internal mapping system. In both cases, the AI is taking in information and making a decision around it.

The more complex things a robot is capable of, the more types of information it has to interpret. It also may be assessing multiple sources of one type of data, such as, in the case of aural data, a live voice, a radio, and the wind.

As robots become more complex and are able to act on a variety of information, it becomes even more important to determine which information the robot acted on, particularly when harm is caused.

Accidents happen

As with any product, things can and do go wrong with robots. Sometimes this is an internal issue, such as the robot not recognising a voice command. Sometimes it’s external – the robot’s sensor was damaged. And sometimes it can be both, such as the robot not being designed to work on carpets and “tripping”. Robot accident investigations must look at all potential causes.

While it may be inconvenient if the robot is damaged when something goes wrong, we are far more concerned when the robot causes harm to, or fails to mitigate harm to, a person. For example, if a bionic arm fails to grasp a hot beverage, knocking it onto the owner; or if a care robot fails to register a distress call when the frail user has fallen.

Why is robot accident investigation different to that of human accidents? Notably, robots don’t have motives. We want to know why a robot made the decision it did based on the particular set of inputs that it had.

In the example of the bionic arm, was it a miscommunication between the user and the hand? Did the robot confuse multiple signals? Lock unexpectedly? In the example of the person falling over, could the robot not “hear” the call for help over a loud fan? Or did it have trouble interpreting the user’s speech?

The black box

Robot accident investigation has a key benefit over human accident investigation: there’s potential for a built-in witness. Commercial aeroplanes have a similar witness: the black box, built to withstand plane crashes and provide information as to why the crash happened. This information is incredibly valuable not only in understanding incidents, but in preventing them from happening again.

As part of RoboTIPS, a project which focuses on responsible innovation for social robots (robots that interact with people), we have created what we call the ethical black box: an internal record of the robot’s inputs and corresponding actions. The ethical black box is designed for each type of robot it inhabits and is built to record all information that the robot acts on. This can be voice, visual, or even brainwave activity.

We are testing the ethical black box on a variety of robots in both laboratory and simulated accident conditions. The aim is that the ethical black box will become standard in robots of all makes and applications.

While data recorded by the ethical black box still needs to be interpreted in the case of an accident, having this data in the first instance is crucial in allowing us to investigate.

The investigation process offers the chance to ensure that the same errors don’t happen twice. The ethical black box is a way not only to build better robots, but to innovate responsibly in an exciting and dynamic field.

Keri Grieman, Research Associate, Department of Computer Science, University of Oxford

This article is republished from The Conversation under a Creative Commons license. Read the original article.

Intelligent robots transforming industry across the world

Nicole Clement - Chief Marketing Officer, Andrew Lloyd - E-Mobility Segment Leader, Giovanni Di Stefano - Head of Innovation and Process Technologies, Comau

In recent decades, robots have become increasingly central to the evolution of a range of industries and industrial processes, chief among which is manufacturing. Advanced robotics is now found in factories all over the world, driving faster and more accurate production of goods.

Soon, they will move out of the factory, for example, for solar panel installation, where robots can really help to increase productivity levels, as well as the assembly of solar blades. Their importance is such that they have been integral to what is commonly referred to as the Fourth Industrial Revolution, or Industry 4.0., where humans are at the centre of the production process and efficiently and safely collaborate with robots and other industrial machines.

But just as any revolution only marks the start of a new era, so too are the robots that brought us to this point only the start of a new phase in the development of intelligent technology. They may have already had a transformative impact on numerous industries, but as demands for more intelligent industrial processes grow, robots themselves are undergoing their own evolution.

The thinkers of tomorrow

Robots have traditionally been ‘doers’: they perform tasks set by humans, according to strict directions given to them. Their go-to function is repetition. What they will need to develop is the capacity to think independently of their human overseers. In short, robots will need to become ‘thinkers’ – they will need to interpret dynamic data and respond accordingly. In doing so, they will be better able to work alongside human operators in a symbiotic relationship where each focuses on complementary value-added activities.

This is necessary for several reasons. Market conditions have evolved as industrial processes in general have grown more sophisticated, and this has changed the nature of competition. In manufacturing, competition over time-to-market for goods has increased dramatically as new technologies have been introduced to the factory floor. This in turn has increased the need to manage and minimise unpredictability in manufacturing processes – to eliminate as much as possible any delays and supply shortages and to perform functions at higher speeds and with heightened accuracy.

Introducing to any operation a robot that can think as well as do – in other words, a robot that is equipped with intelligent automation functions – satisfies the dual need for efficiency and reliability. It enables better resource use and increased productivity while simultaneously minimising the likelihood of error, improving quality as a result.

Transforming industry, robot by robot

Comau has been a global leader in the field of industrial automation for close to half a century. Thanks to its continuous commitment to innovation, research and advanced training, it offers its customers expertise and new technologies in the most progressive sectors, including digital transformation and electrification.

The company, which is headquartered in Turin, Italy, understood early on that manufacturing would be transformed were robots to become more than just doers, and it set about developing advanced Industry 4.0-enabled robots that could draw on the latest developments in AI and IoT to enhance their performance. AI streamlines deployment and removes the costs for time spent programming and testing, while IoT features put the robot ‘in touch’ with everything in its environment and predicts where maintenance is needed in advance of obstructions or equipment breakdown. Lasers and AI vision systems can also ‘see’ potential problems up ahead and thereby reduce downtime.

The innovations it has introduced are transformative, and their application is far-reaching. Comau’s robots and technologies are present in every industry across the globe as one among only a small handful of companies that have such a wide-ranging offering.

Its robots meet two pressing needs of industry in the modern day: variability and customisation. These features are central to the ‘thinking’ robot. Rather than stopping when they encounter an error – say, an engine part in the wrong place on a factory line – the robots are now equipped to look for errors or imperfections themselves and respond immediately, thereby allowing them to continue with their task without delay. And just as they can use the powers of IoT to sense what is around them and predict the actions of nearby machines, so too are those nearby machines now able to do the same of robots. The whole ‘smart’ factory environment is in a state of symbiosis, and this has a profound effect on efficiency and safety.

Comau has also made sure that the use of these robots, together with a variety of other cutting-edge technologies, is more democratic than it has been historically. The company’s aim all along has been to provide not just for the giants of industry but also for small- and medium-sized manufacturers, regardless of what they produce. Its solutions, in the form of hardware, software, vision systems, AGVs and more, are all developed in-house and are designed to respond to the unique needs of each customer. It works side by side with the customer throughout the entire process of acquiring, installing and testing its robots.

Freedom to think

The transition from industrial to intelligent robotics is already transforming industrial processes and is set to do so well into the future. Designed into the robots of today is a sense of freedom and autonomy – they now have a level of intelligence that means they can make their own decisions, free of human interference. This in turn frees up human labour to focus on the jobs robots cannot do.

Comau sees the future as one of endless enhancement of robots’ capabilities. They will be faster and lighter; they will be completely mobile and able to go everywhere; they will play a central role in sustainability; and they will work better alongside people. In short, their impact on industry, as well as the work-life quality, will be game-changing.

For more information please visit www.comau.com

INDUSTRY VIEW FROM COMAU

How to maximise your AI investment

Wolf Ruzicka, Chairman, EastBanc Technologies

Artificial intelligence (AI) can be found almost everywhere in modern life. Whether you’re receiving financial advice via a banking app, shopping online or troubleshooting computer problems with tech support, AI is likely to play a part in your daily activities. Indeed, 50 per cent of respondents to McKinsey’s State of AI 2020 survey reported that their companies have adopted the technology in at least one business function.

With the size of the global AI market valued at more than $60 billion in 2020, it’s clear that an incredible amount of money is being poured into this technology. However, companies should be wary of the misconception that AI in and of itself will deliver a return on investment (ROI). With such widespread adoption, key lessons and best practices are emerging to help companies avoid common AI pitfalls and achieve ROI from their AI systems.

Most common AI mistakes

AI is not a switch companies can simply "flip on", and there’s no one-size-fits-all AI plug-in. Despite big investments and seemingly expert advice from knowledgeable vendors, many companies still make mistakes along the way. These mistakes have both tangible and intangible consequences, including loss of sales, unnecessary costs and, perhaps most importantly, a loss of end-user trust. Here are some of the most common mistakes made when deploying AI systems, and how best to avoid them:

Insufficient penetration: AI is more complicated to implement correctly than many companies realise at the outset. Designed to be part of a holistic business system, it will offer little benefit if only installed at a surface level. For example, many companies use AI in a chatbot function on their frontend. As these bots typically don’t have access to a company’s core systems, they are no help beyond the most basic of functions and are easily identified as non-human by customers. Without access to the right datasets on the backend, this use of AI will fail to make a meaningful impact on a company’s bottom line.

Incompatibility between different AI systems: Even those businesses that have incorporated AI into their core systems still aren’t guaranteed meaningful ROI. A company could be running multiple AI engines at once to support multiple business functions. Problems occur when these engines don’t communicate with each other effectively, or give conflicting results and advice.

Inability to go big: Small-scale AI will only offer small-scale returns. The inability to roll out the technology on a large enough scale holds many companies back from reaping the rewards of their investment. Interestingly, it’s often big organisations with unwieldy backends that struggle with this the most.

Vendor bias

Vendor bias is another reason why many organisations fail to get their money's worth after investing in AI. Companies traditionally outsource the entire job to a single vendor that delivers an end-to-end solution. However, such huge and abrupt system overhauls are costly, slow and very risky. Most pertinently, this approach also leaves the company with no control or autonomy over the systems they come to rely on everyday. Vendors also naturally prioritise their own technologies, meaning that the vast majority of products on the market are excluded, even if they would provide the best solution for clients.

In contrast, thanks to a robust AI ecosystem, companies can select best-in-class products that can be implemented in a seamless and modular fashion to meet their unique needs. You can’t just set it and forget it when it comes to AI. Systems should be flexible and adaptable to incorporate the best that today’s rapidly changing market has to offer.

“Ultimately, what you really need to understand is that the core of this problem lies in the core of your business, not the technology vendor’s business,” says Wolf Ruzicka, Chairman of EastBanc Technologies, which helps companies customise and better leverage their existing AI systems. “Instead of having this technology bias, you must own up to the fact that you need to own your own technology destiny."

The solution

Only the company itself can drive a modular custom approach that perfectly complements its unique goals, value proposition and customer needs. But most companies don’t have this skillset within their existing talent pools. That’s where EastBanc Technologies steps in.

With more than 20 years of experience, the Washington DC-based team of software engineers puts its clients in the driver’s seat by enabling them to design, build and own their AI systems. Supporting and empowering every step of the way, the EastBanc Technologies team helps companies build modular custom software that quickly unblocks problems and delivers impactful returns.

The EastBanc Technologies team starts by identifying a “killer feature” – the unique selling point at the core of the business model that draws the end-user in and evokes emotion. Once the killer feature is identified, an AI module is integrated to enhance that feature. When this first feature is working as it should, other business applications and functions are brought online around the killer feature, progressively cleaning up and connecting data streams throughout the business to the AI systems. Unlike the traditional model, this incremental approach prioritises organic permeation of AI. It is a fast, flexible and low-risk approach that's laser-focused on ROI.

“All that companies really have to do is commit to not outsourcing this fundamental addition to their business,” says Ruzicka. “[They can add] components gradually on a very granular level to become AI leaders in their respective spaces.”

For more information on EastBanc Technologies and its AI services and solutions, watch the video above or visit the company’s website.

INDUSTRY VIEW FROM EASTBANC TECHNOLOGIES

How viable is citizen development?

There is now a $20 billion market for no-code apps and the citizen developer movement is at the peak of inflated expectations. How did this happen?

Citizen development is promoted as one way to put the power back into business users’ hands. The movement took its position on account of the amount of shadow IT (anything that can’t be monitored or managed with full transparency by an IT department) that has grown in large corporates over the past 20 years. We only have to look back at the Covid crisis in the UK to understand the risks inherent in shadow IT – an Excel spreadsheet error was blamed for 16,000 missing coronavirus cases.

With the advent of RPA, a proliferation of low-code and no-code tools exploded on to the market. These easier-to-use tools took advantage of technology that was too slow to change or too rigid to respond to business needs. But we may still wonder if it is that simple. Many IT execs and specialists have a wry smile about the viability of a citizen developer approach, with the word ‘developer’ usually sparking some concerns. Can the business really code without any background in coding? Perhaps instead, the question we should be asking is: can the business get by without it?

There is all too often a shortage of IT talent and budget for business leads to manage processes effectively. In the banking domain, a lot of banks’ internal systems are now decades old, problematic to expose APIs and slow to make changes. Operations users in the back office need flexibility to log and track comments on why something is booked the way it is and a tool that allows them to manage their operations without having to take on a headcount of hundreds.

Low-code/no-code tools can offer support for multiple file types and visibility of data enrichments without code hidden in the back end. Most importantly, there is less need for a developer to build a reconciliation. Already, this low-code configuration means change management timelines can reduce from months to days – a crucial differentiator when regulations can so rapidly change.

Low-code cases

To provide regulators with more transparency in real-world banking operations, business users already use some auditable tools, but they are often neither intuitive nor have user-friendly UIs. Other reconciliation tools in use have sophisticated code that does not make sense to business users. This is a key issue, since the way a reconciliation has been set up needs to be explainable to the regulators by the business. Likewise, if the builder leaves the company, there is often no traceability requirements for the reconciliations they have built?

From this lens, the frustrations of the business become clear. Key challenges that banking reconciliations have to face are cross-departmental reconciliations taking considerable time and effort, audit gaps due to the use of Excel, and the manual nature of adapting to regulation changes through data mapping.

Once the bank has upskilled the business in the relevant tools and is past the experimentation phase, the hard and soft benefits of citizen development are multi-fold. In one scenario, timelines to run a reconciliation were reduced by 50 per cent and there was a substantial reduction in licence costs, with some tools completely decommissioned. Furthermore, the bank adopted a simple and easy-to-use tool for non-technical SMEs, which brought further employee satisfaction and engagement.

As organisations have become more confident in digital offerings, citizen development is currently on the slope of enlightenment. But let’s be clear: these are simple and functional tools rather than highly innovative ones. Business operational leads don’t need a tool of mass innovation – they need a tool of mass usability to help them transition to the plateau of productivity that citizen development offers.

INDUSTRY VIEW FROM ZUKUNFT

Relieving CSRs from repetitive manual work with attended automation

Oded Karev, General Manager, NICE RPA

For contact centre agents, the past couple of years have certainly not been easy. That’s why leading companies are looking at how they can use automation to address the pressures agents have experienced working from home and dealing with massive call volumes from distressed customers seeking reassurance from a human voice.

The new pressures put further strain on the stress points that have always existed in call centres: the challenges of keeping staff up to date with ever-changing tech and regulation, meeting higher customer expectations and driving higher efficiencies. These challenges actually lead to lower employee engagement and motivation, while increasing the need for retention as many employees are experiencing burnout.

Many agents are frustrated with the technology and processes they use to do their work as well as with the work experience itself. The US Contact Center Decision-Makers Guide 2021 reveals that 81 per cent of contact centre decision-makers agree that multiple copy-and-paste leads to wasted time and errors.

Another 68 per cent said that it’s important to reduce after-call work, while 76 per cent agreed that agents find it difficult to learn new systems. Perhaps most concerning of all is the finding that repetitive work remains among the top three reasons for agent attrition.

The good news is that contact centre decision-makers are recognising the role that digital technology can play in resolving some of these challenges. Seven in ten, for instance, report that robotic process automation (RPA) can help to reduce average call handling times.

Attended automation: the next step forward for call centres

The benefits of unattended RPA in the back-office are, of course, well understood now. They reduce costs, enable scalable operations and let people focus on work that requires strategy, creativity and interpersonal skill rather than on repetitive processes.

Leading organisations are looking at taking automation a step further by putting robots on the frontline of customer service as enablers for contact centre agents. These desktop robots, or attended automation, can assist agents to perform efficiently and accurately by taking away the need to manually navigate multiple screens and apps.

Attended and unattended robots working together

The benefits of automation really begin to compound when attended and unattended process bots are blended to scale operations and drive higher efficiencies. For example, an attended bot could automatically populate a form or provide the agent with links to data and real-time next best-action guidance as they help a customer to open a new bank account.

Unattended bots could be used to generate an email to the client after the call with the agent is complete, or to generate and categorise technical support tickets on behalf of the service agents. This combination of attended and unattended technology lets people focus on adding value rather than on processes and systems.

For automation to be successful and sustainable in a contact centre environment, it needs to enable agents in real time. Today’s sophisticated blend of cognitive, attended and unattended automation solutions delivers this functionality, helping to bridge the gap between employee engagement, customer experience and cost containment.

Today’s leading call centre operations are leveraging the digital workforce not only to make human agents more productive, but also to give them a better employee experience and to empower them to deliver a better customer experience in turn. This approach satisfies many of the contact centre agent’s core motivations and skills:

The power to help the customer
Getting a 360° view of the customer in a single view
Research support
Guidance around compliance
Ability to work efficiently to help each customer and serve more people

Enabling contact centres to scale and perform

Most call centres want to drive significant and continuous improvement in six ways:

Increasing customer satisfaction
Reducing handling time and increase efficiency
Increasing first-contact resolution
Reducing training time
Improving compliance and process adherence
Increasing upselling during service calls

Automation enables enterprises to achieve these goals.

At NICE, we have a long history in contact centres as well as a strong track record in attended and unattended automation. Our products are built on 20 years of deep understanding of contact centre operations and technology – and we can offer an integrated suite of RPA and contact centre solutions.

With NEVA, our personal agent assistant, we have pioneered attended automation. NEVA resides on the agent’s desktop, helping them in real time and in a contextually relevant manner via interactive callout screens. Her unique capabilities ensure real-time process optimisation and automation of desktop tasks, resulting in improved employee and customer experiences.

Our ability to span the back-office and the contact centre with a comprehensive, intelligent automation solution is unique in the market. We would welcome an opportunity to discuss how we can help your contact center reconcile the customer experience and employee engagement challenges it faces today.

Contact us to learn more.

INDUSTRY VIEW FROM NICE

3D printing offers African countries an advantage in manufacturing

Thousands of years ago, the blacksmith led a technological leap in sub-Saharan Africa. West Africa’s Nok culture, for example, switched from using stone tools to iron around 1500BC. Imagine an innovative artisan like this re-emerging in the 21st century equipped with digital technologies.

This is not Wakanda science fiction. It is the story of a real promise that 3D printing holds for an industrial revolution on the African continent.

3D printing, also known as additive manufacturing, is a fabrication process in which a three-dimensional object is built (printed) by adding layer upon layer of materials to a series of shapes. The material can be metal, alloys, plastics or concrete. The market size of 3D printing was valued at US$13.78 billion in 2020, and is expected to grow at an annual rate of 21 per cent to a value of US$62.79 billion in 2028.

Not only is it a different way of physically making objects, 3D printing also changes the picture of who can participate in industry – and succeed.

3D printing is an excellent match for smaller operators because it does not require huge capital investment. It is also the best fit for “newcomers” while established operators are locked in the old manufacturing method. The new technology is a great opportunity for developing countries to leapfrog over developed countries.

In a recent paper, we reviewed the evolution of 3D printing technologies, their disruptive impact on traditional supply chains and the global expansion of the 3D printing market.

We show that conditions in the African context are favourable for technological leapfrogging, and propose that universities, industries and government can work together to support this, giving small and medium enterprises a key role.

We illustrate our argument using South Africa and Kenya as examples.

Technological leapfrogging

Technological leapfrogging is related to technology lock-in.

Lock-in happens when an established technology continues to dominate the market even after the arrival of a new and superior technology. The older technology remains successful not because it’s better but because it got the advantages of an early lead in the market.

In developed countries, where the older technology has taken hold, it’s difficult for new, radical technologies to get a start. Too much has already been invested in the old ways.

But it’s different in developing countries. Less has been invested in older technologies. And almost everyone is starting from the same point; the cell phone is an example.

For a long time, the African continent has lagged behind the rest of the world in manufacturing. A recent report indicates that while Africa is home to 17 per cent of the world’s population, it accounts for only 2 per cent of global manufacturing value added. 3D printing presents an opportunity to revive this sector through technological leapfrogging.

African countries meet the four key conditions highlighted by scholars for technological leapfrogging:

There must be a large difference between the wage costs of the leading nation and potential challengers.
The new technology must appear initially unproductive and less profitable relative to the old.
Experience in the old technology must be less useful and less transferable to the new technology.
The new technology must, in the long run, improve productivity and efficiency.

To take the first condition, the wage cost of an average African country is a small fraction of the wage cost in a developed country. For example, according to the latest estimates, the average annual income in Nigeria is US$2,000, compared with US$64,530 for the United States.

3D printing is initially unproductive because of lower initial rates of adoption. This means a smaller market and limited profit opportunities.

Looking at the third condition, 3D printing is not an incremental improvement on what went before, so experience in the old technology does not count for much.

On the fourth condition, one of the strongest arguments for 3D printing is that it flips the dominant logic of traditional manufacturing: scale economies. Big multinational manufacturing corporations invest heavily in machinery, logistics and other material and human resources for mass production. They make big profits only if they sell enough units. The more they sell, the bigger their profit margins.

3D printing doesn’t need centralised high-volume production and large inventory stocking. Suddenly, it pays to produce fewer units. There is no need for heavy investment in manufacturing plants, because 3D printers come in various smaller sizes and at lower costs. There is now a growing market for budget and do-it-yourself 3D printers that cost less than US$200.

Smaller and more sustainable

All this shifts the advantage in favour of micro, small, and medium scale enterprises.

Firstly, it offers greater reward for creativity and ingenuity. Like the African blacksmith of yore, additive manufacturers can design customised, higher value products in response to specific demands and requirements.

The proximity of 3D printing shops to customers is another advantage as it reduces logistics costs and supply chain challenges.

Sustainability is another benefit: the process produces only what is needed. It can reuse waste material.

Micro and small-scale 3D printing shops can offer work and income opportunities for households.

University, industry and government

Our study proposes a way for the university, industry and government sectors on the African continent to work together to harness the opportunities offered by 3D printing. These domains – producing knowledge, producing goods and regulating economic relations – have tended to be disconnected. Instead, we argue that greater integration can encourage innovation.

We give examples from South Africa and Kenya to illustrate the challenges and opportunities.

In South Africa, universities are leading the drive to provide training and retraining programmes for engineers, technologists and other professionals involved in 3D printing. Much more needs to be done to develop new curricula, research and programmes in additive manufacturing.

Kenyan universities are at an earlier stage, focusing on convening networking and knowledge exchange events.

In the government sector, South Africa has the most detailed policy document of any African country on 3D printing. The country’s 3D printing strategy is being led through the Ministry of Science and Technology, and through agencies such as the Council for Scientific and Industrial Research and Technology Innovation Agency. In the industry sector, South Africa’s Rapid Product Development Association works closely with the government to organise conferences, workshops and community engagement activities.

The results so far

The South African 3D printing industry has had considerable success in recent years, driven by a growing community of enthusiasts and designers.

Small enterprises and startups are making inroads in areas such as 3D printing of cell phone accessories, car accessories, and jewellery. In 2014, South African doctors used 3D-printed titanium bones to perform a jaw-bone transplant surgery, the second in the world. There are also recent applications of 3D printing in housing.

The three spheres need to do more work in research investment, policy interventions and strategic public procurement. And they need to cross boundaries. Universities can commercialise and contribute to policies. Industry can invest in research and influence policies. Governments can play in the market and in knowledge production.

Seun Kolade, Associate professor, De Montfort University

This article is republished from The Conversation under a Creative Commons license. Read the original article.

The smarter way towards smart factories

Michael Goepfarth, CEO, SCIO Automation

Despite a surge in interest over the past decade, automation isn’t new – and nor is the concept of the “smart factory”. Manufacturers and designers have been trying to make factories “smarter” since the late 18th century, when new manufacturing processes developed in Europe gave rise to the industrial revolution and brought to the production of goods a level of speed and efficiency that had eluded previous generations of workers.

But the automation of factories today of course has a very different feel to that of the Industrial Revolution. By the late 1700s, new machines were appearing on the factory floor, but large workforces were still indispensable to the manufacturing process. There were no computers that could gather data on output, or that could predict where problems in the factory line might occur. There was heavy wastage, both material and financial, and production – despite being much faster than before – was still slow.

In the present era, however, innovations in technology stand to drastically reduce waste and enhance efficiency. The “smart factories” currently being imagined and developed are fully digitised spaces where AI and machine learning provide fine-grained data on every aspect of the operation. They have come about in tandem with advancements in logistics, among them smart warehouses and conveyor systems, and autonomous material flow – developments that are essential to the “smart” functioning of factories.

The myriad aspects of the manufacturing universe – the factory, the supply chain; even the talent pipeline – are intimately connected through new technologies; indeed, the processes that have improved production have grown together with those that improve logistics, such that the two have a symbiotic relationship. In the factory, floor managers can see in real-time how efficient production is on any given day, where wastage is occurring, and where problems, such as blockages in a machine, might arise. So too is each component of a supply chain – where transportation of goods might be problematic; where supply might be low – subject to far closer scrutiny than previously.

In short, the smart factory of today, and the environment it sits within, brings a far more enhanced level of operation than previous manufacturers enjoyed. But not everyone is jumping on board. Manufacturing companies can be unwieldy things, averse to change and lacking the flexibility to take their operations in a new direction. Yet they also know that if they don’t evolve, they will lose out on business, and costs will remain high.

SCIO Automation understands that technical innovation should no longer be viewed as merely an advantage – something that will make a business stand out from the crowd – but as key to the survival of companies involved in, or reliant on, manufacturing. It’s not a niche product; rather, it has become central to competitiveness. Without modern automation, businesses will die. SCIO has been tailoring integrated automation needs at all operational and information technology levels to clients for decades, and in the process has become a linchpin of the global transition to Industry 4.0. If there isn’t a solution for its clients currently on the market – for instance, Autonomous Mobile Robot solutions for the demanding and flexible transportation of parts from the warehouse shelf to the production line – SCIO can design one.

Central to smart factories is the harmonising of the different logistical and production elements of their operation. Every aspect needs to work together, towards one goal. But companies have struggled with the transition to smart manufacturing in part because they have taken automation on a component-by-component basis. SCIO impresses upon clients the need for the connectedness and symbiosis of logistics and production, and within that, all the “smart” parts – sensors, connected devices, cloud computing, Big Data and more. This is important because improved productivity and efficiency rests on managers being able to see the entire operation in real time, and in one place. Huge advancements have been made in the technologies designed to improve efficiency on the factory floor, as well as along the supply chains that products pass through when they leave the factory. Not every business has the confidence to make the changes necessary to stay competitive, however – these changes are, after all, daunting, and there is risk. But with the right kind of help, those risks can be mitigated. And when manufacturing businesses do begin their march towards Industry 4.0, the long-term benefits of smart factory technology – improved workplace safety, enhanced productivity, minimised waste, lower operating costs – will soon materialise.

Get more information about SCIO at www.scio-automation.com.

INDUSTRY VIEW FROM SCIO AUTOMATION

The personification of the black box

Dr Sam Anthony, CTO and Co-Founder, Perceptive Automata

Autonomous vehicle progress is stalling because AI sees pedestrians as black boxes. But there is a way to solve it

When you use traditional AI techniques in a self-driving car, you end up with a vehicle that sees humans as black boxes. These boxes move around, and sometimes you can attach labels to them – this one is tall, this one is small, this one is holding its arm up – but you don’t understand them.

Navigating around black boxes is hard. They move in unexpected directions, and if you don’t want to hit them you have to be incredibly cautious, assuming they could move in any direction at any time. In fact, there are many situations where it’s simply impossible to figure out how you can get past a black box if you can’t hit it.

It would be better if you could understand them not as black boxes, but as people – people who have ideas and goals and are trying to figure out how to interact with you.

A big problem for autonomous driving systems

When traditional AI models are trained, what they’re learning is a mapping of a label to an image. You take thousands of images and generate numbers for them. How those numbers get generated, or what those numbers mean, isn’t part of the process, but the AI learns what images tend to go with what numbers. What the AI is capable of doing is predicting what the black box attached to that image should say and do. It’s trying to figure out which image is represented by which black box. That’s it.

When you put that system in a car and it sees a pedestrian out in the world, it matches that pedestrian with the black box that fits it best. But it doesn’t really know anything about that pedestrian. It doesn’t have any ability to reason about what’s in that person’s head. Black boxes don’t want to cross the street – black boxes don’t want anything. Black boxes don’t know you’re there. They have no inner life.

Solving the problem with research

At Perceptive Automata, we remove the black box. When we train AI, we do something different. We still take thousands of images, but instead of opaquely, mindlessly applying labels to them, we integrate the personhood of the labellers deeply into the training process. As each of those thousands of images is shown to the AI, what it’s learning is not a set of disconnected numbers. It’s learning what people think about that image. In particular, it’s learning how people would answer questions about what’s in the head of a pedestrian pictured before them.

This isn’t easy to do. To ask people questions about what’s in the minds of pedestrians and get answers that are usable for training AI requires scientific rigour and a great deal of art. You need expertise in visual psychophysics, a field of science dedicated to measuring how people see and respond to the world. You have to be a people expert and know how to understand, study and characterise them.

AI trained this way, like Perceptive Automata’s SOMAI, or State of Mind AI, no longer sees pedestrians as black boxes. When you put our AI in a car, it sees pedestrians as people do. Instead of merely identifying what set of numbers maps most accurately to a given pedestrian, SOMAI is able to answer questions about what’s in that pedestrian’s head. It does this by imagining the people who trained it and hearing their voices. In effect, it answers the question: “If there were 500 people here in the car with you, and you asked them all, what would they say about whether that pedestrian wants to cross in front of your car?”

When you understand pedestrians as people, who have goals and desires that interact with the goals of the vehicle, they stop being black boxes that might move anywhere at any time. They want specific things. Some of them standing at a crossing want to cross. Others maybe don’t. Without a real-time understanding of what’s in pedestrians’ heads, autonomous vehicles will be stuck trying to navigate around black boxes, and the promise of this industry will not pay off.

Learn more and request a demo of SOMAI at perceptiveautomata.com

INDUSTRY VIEW FROM PERCEPTIVE AUTOMATA

Embracing IoT in UK business

If organisations look beyond initial barriers to IoT adoption, they can unlock a new realm of improved productivity and efficiency

Maintenance issues can quite accurately make or break business operations. For manufacturers, this is the reason why utilising the ever-growing IoT means they are able to accelerate smart manufacturing and rapidly improve their processes. IoT can help augment a number of different business operations, from customer experience to process maintenance. In manufacturing, artificial intelligence and IoT applications can efficiently deal with various operations, from predictive monitoring and preventative maintenance through to optimising equipment performance, quality control in production and even the much talked about human-to-machine interaction. All of this encompasses a reduction in product cycle time and greater efficiency through reduced downtime.

So why is it that adoption of these technologies is slower in the UK?

There is an argument that it is the public sector that is causing the backlash of IoT adoption, due to a requirement to provide a strong return on investment. But in the manufacturing industry, the downtime reduction alone is a clear return.

One good theory is the idea of lack of infrastructure throughout the UK. IoT adoption required brilliant communication systems, fast speed, reliable networks and, in many cases, 5G connectivity with its boasted minimal latency. However, it doesn’t appear that sufficient investment has been made to support these areas.

Throughout the UK, the most significant IoT advances have been made by both the healthcare and military sectors. Through patient monitoring and remote monitoring hospital check-ups, the NHS is becoming a pioneer for this technology.

Another barrier to entry sometimes can be the time lag in visualising the return. Many IoT development projects can sometimes take time and involve workforce upskilling and higher initial costs. On average, IoT projects take around two to five years, and for many organisations this is a tough investment to make. However, this can be down to the lack of expertise and confidence among executives and board members who are the decision-makers for these technologies. Often, it is the person on the shop floor who can really feel the benefit from these technologies, but they are not the decision-maker – causing a lack of connection.

It is essential that the industry is reassured that using IoT will most certainly improve processes, and the speed at which it is developing is now even reducing the security vulnerability factors. Another concern from many businesses surrounds privacy and security constraints.

To ensure more projects deliver the benefits they are looking for, UK organisations need to plan ahead to address current and future challenges. Connectivity must be meticulously considered from the get-go to ensure that issues along the way don’t impede the ultimate rollout.

It is essential that upper management and executives understand the impact of high device volumes from a cost and resource perspective, ensuring greater visibility across security, maintenance and performance monitoring. They must seek to create a more efficient working environment for their workforce through these technologies in order to successfully realise a return on investment.

Of course, along with infrastructure, an investment in IoT security will become increasingly critical and must also be understood to be as reliable as older security systems.

Overall, the UK is continuing to build momentum around the IoT. Yet there is still much to do to progress to faster, wider, large-scale adoption. If organisations are able to understand that maximising the value of IoT-generated data will help future-proof projects, growth can be unparalleled.

To learn what other UK organisations are doing on their journey through IoT technologies, visit gambica.org.uk

by Nikesh Mistry, Sector Head – Industrial Automation, Gambica

How to make more time and money from your manufacturing operation

2020 will see a further increase in the deployment of industrial digital technologies within UK manufacturing operations. These technologies can help manufacturers address some of their pain-points and create new gains for their customers, shareholders and workers.

Let’s face it, there’s been no shortage of excitement or promotion around so-called Industry 4.0 technologies such as the internet of things, robotics and automation, machine learning, 3D printing, artificial intelligence and augmented reality.

Cutting through all the jargon, we at the Institution of Engineering and Technology (the IET) would like to de-mystify some of the hype that beckons you to jump on the “digital bandwagon”, particularly if you are a small or medium-sized enterprise owner, manager or investor.

How will any of this improve the things that really matter?

At any one time, there are a myriad of issues facing SME manufacturers, many of them completely beyond your control. The challenges are many, varied and specific to each firm and its niche or sector. And it’s no secret that with challenges come opportunities too!

• Lack of visibility

• Skills and staff shortages

• Fulfilling customer orders

• Rising costs, such as energy

• Purchasing/inventory

• Productivity improvements

• Product quality/consistency

• Machine downtime

• Power outages

• Legacy premises and old equipment not fit for purpose

• Retaining existing customers

• Getting paid on time

• Keeping a constant flow

• Prototyping costs and time

• Too much time spent firefighting

• Winning new orders

• Time to market taking too long

• Matching capacity to demand

• Limited funds for CAPEX

• Supply chain issues

Industry 4.0 technologies won’t necessarily solve any of these issues for you outright, but they will enable you to hone in on and quantify solutions to those things you can directly inspire, inform and influence. Harvesting, analysing and acting on the right data in real time offers increased speed and ability to address your pain points within the business and lies at the very heart of Industry 4.0.

Why should I even spend time thinking about all this?

Fundamentally, there are two reasons. First, reduced costs. Your operating costs should fall and your available time should rise as a result of using the right digital tools within your business.

Second, that you should stay ahead. It’s likely that many of your competitors, collaborators and clients may well be exploring or increasing their use of digital technologies within their businesses. Stay in the game, get yourself up to speed and avoid getting left behind by innovating before they do.

Where do I start?

Set your sights high but start with a grounded view. Don’t spend money on “digital”, if you haven’t already optimised your “physical”. The adage remains: get lean, then get digital. You need to find out what’s really happening within your manufacturing operation, or as we say, create a single version of the truth. To do this you will need to digitally connect your existing machines and information systems across the business.

This used to be the privilege of big businesses that could afford expensive bespoke programmes to connect their systems. The new digital tools bring such connectivity between systems such as ERP (enterprise resource planning) and CRM (customer relationship management) within the grasp of any SME.

To complete this task, it’s likely that you will need to add some simple and relatively inexpensive sensors to your existing machines (at the cost of a few pounds) and some new connecting protocols to your network.

To do this and make sense of the data generated, you may need to get help. Challenge your new apprentices or latest recruits to work with your champion on this. Failing that, try contacting your local further education college, university engineering department, equipment supplier or catapult centre.

What next?

Having gained a better understanding of the key factors at play within the business, you’ll be in a much better position to shine the spotlight on those parts of your operation which require deeper examination, and that will give you savings and increased flexibility. It’s vital to act on these insights of your operation and reap the rewards before moving forward to the more advanced steps where you will need to invest your hard-earned cash on further technology.

As anyone who has ever been through a new ERP or control system implementation knows, there is no point at all in digitising poor productivity (at best) or digitising chaos (at worst).

Creating new gains

Industry 4.0 is all about taking your existing human capital, shop floor equipment and back office systems and connecting these valuable assets, giving you a clearer and faster view of your world, and enabling your team to save money and time, invest your savings in the right technology at the right time with clear return on investment, and spend more time with your existing and new customers to grow your business.

by John Patsavellas, Senior Lecturer, Cranfield University and expert panel member at the Institution of Engineering and Technology

The climate crisis: ensuring positive change

John Riggs, SVP Applied Technology Solutions, HSB

In April 1865, two weeks after Abraham Lincoln was assassinated, an overcrowded paddle-steamer exploded on the Mississippi river just outside Memphis. Its overworked, badly patched steam boilers were to blame. More than 1,500 people – mainly Union prisoners being transported home from the US Civil War – lost their lives.

To this day, the disaster that befell the SS Sultana remains the worst in US maritime history. It wasn’t a one-off: steam, which had become an indispensable element of the Industrial Revolution, and the central power source for a wide variety of functions, including transportation, manufacturing and heating, was dangerous and unpredictable. Steam boilers had a tendency to explode, and when they did the results were catastrophic.

A group of engineers in Hartford, Connecticut took up the challenge: how do you make a safer steam boiler? After much strenuous experimentation, they landed on a set of technological innovations they could back up with guarantees. HSB (Hartford Steam Boiler) was born.

This, of course, provides a too-close-for-comfort metaphor for our current predicament. We are all passengers on the SS Sultana, except this time around it isn’t a steam-paddler – it’s our planet. And there aren’t a bunch of steam boilers at risk of exploding. Rather, there are myriad deeply intertwined natural systems that are overheating, breaking down or on the verge of blowing up. The evidence is abundant wherever we look.

What can we, as insurers, do about this? The immediate and somewhat obvious response is to offer more robust insurance, so that when the climate-related flood, fire or windstorm inflicts catastrophic damage on your property, we’ll help to make you whole again. This sort of response is all well and good, but only addresses the consequences of climate change.

We’re much more interested in helping to slow, even reverse, climate change. There are two areas where we can make a difference:

- Vastly increasing the efficiency of the currently installed base of industrial equipment, thereby lessening energy consumption
- Converting, where possible, existing equipment to renewable fuels and ensuring that, wherever feasible, new equipment and machinery run only on renewables

The value we bring relies on a number of closely related elements:

- We partner with best-in-class OEMs and operations providers
- The data and analytics we bring to bear via the industrial internet of things (IIoT) and AI hugely multiply the visibility we have into the performance and fine-tuning of the projects we underwrite
- We write many of the standards. Our benchmarking abilities are the envy of the industry
- We take on the risk by providing innovative guarantees (for example, “cloudy day” insurance for solar farm operators that makes up shortfalls in output)
- Innovative financing, including “as-a-service” models

Every project is different. Sometimes we’re dealing with large inventories of legacy machinery; others we approach with inventive business models enabled by the enormous progress in sensoring and monitoring techniques.

Let’s be straight: these are enormous challenges, orders of magnitude greater than perhaps anything we’ve ever faced. After two centuries of spewing massive volumes of carbon and other noxious substances into the atmosphere, basing our entire industrial infrastructure – and the wealth it generates – on the burning of the fossil fuels, we are talking about nothing less than the total overhaul of the world’s technological, industrial and agricultural operating systems. How much will it cost? According to Goldman Sachs, somewhere around a staggering $100 trillion over the next 20 to 30 years.

We feel fortunate in having an established track record of applying latest-generation technology to thorny problems, and backing these solutions with increasingly sophisticated guarantees that promote their adoption and reassure our customers. That’s what we, as insurers, are supposed to do, right?

But we don’t have all the answers – far from it. And we can’t do this alone. As a player in a highly competitive arena, we routinely keep things close to the vest. We never show our hand. But given the nature of this set of challenges, let us say clearly and unequivocally: we hope and trust that our peers and competitors are working on this as hard as we are. We hope that they are inspiring their workforces to apply their best thinking to come up with solutions that work.

There is an enormous amount to be solved, a tsunami of work to be done. Those who succeed will be amply rewarded, both financially and – more importantly – in the knowledge that they’re helping ensure the survival of future generations.

We’re in this to win. For all of us.

Full steam ahead!

Read more about our position on climate change and sustainability.

INDUSTRY VIEW FROM HSB

Supercharging customer experience for financial services with Voice AI technology

Kun Wu, Co-founder and Managing Director, AI Rudder

Customer experience will never go back to how it was pre-Covid. The pandemic has accelerated digital adoption and changed consumption habits, from on-demand video streaming, e-commerce and food delivery to more digitalised services such as digital banking and payments. Almost everything is available at consumers’ fingertips as 24/7 availability becomes the new normal.

The rise of a hyper-connected and hyper-convenient world has also led to an exponential rise in call volumes at customer service centres across geographies and industries. As a result, businesses struggle to keep pace as pandemic-weary consumers run out of patience and want both speed and flexibility in their digital interactions.

While some of the world’s biggest brands have taken innovative approaches to meet this new standard of customer service, responding to new customer expectations would mean a radical departure from the status quo.

Faster, more intelligent on-demand customer support

In banking and financial services, customer support functions are oversubscribed. However, things would look very different if your top call centre representatives could work 24/7 without fail. AI Rudder can make this vision a reality through advanced Voice AI technology.

Our Voice AI uses automatic speech recognition (ASR) and natural language understanding (NLU) to process human conversations. Our machines can receive and interpret customer intent in voice communications. Not only that, they can respond and communicate on a near-human-seeming level of intelligence.

From payment reminders to debt collection to quality assurance, Voice AI assistants can take every repetitive task off your customer support’s workload, thus freeing them up to focus on more complex conversations that require a human touch.

Bringing the human touch with AI-powered voice automation

Adopting AI does not need to come at the expense of human relationships. While customers may feel sceptical about automated solutions using AI, the technology helps bridge fundamental gaps in delivering exceptional CX.

A Voice AI solution can help companies extract precious data from incoming customer calls and existing recorded conversations. This data can provide banking and financial services companies with valuable insights that would otherwise be overlooked, and which can even help forecast peak demand periods based on identified patterns. Insights from AI-decoded conversations can also lead to product innovations and service optimisation, making your company stand out in a saturated market.

Customer experience teams can even use real-time data from Voice AI to predict issues before they occur, such as a sophisticated phishing scam targeted at customers. By alerting your customers of potential threats, you’re building greater confidence that their assets are safe in your hands.

Besides improving customer experience, businesses can use AI-driven insights from voice conversations to identify gaps in knowledge across client-facing teams. This precious information can feed into training programs, helping you build a top-class customer service department.

Working with banks and financial services companies around the world

Founded in 2019, AI Rudder develops advanced voice AI technology to help businesses solve B2C communication challenges. We work across various industries, including banking and finance, fintech, insurance and e-commerce.

More than 200 companies around the world use our platform today to augment their human agents with our AI voice assistants, maximising profits, efficiency, and scalability.

With our Voice AI solution, businesses can increase the scale, speed and quality of their customer experience while reducing operational costs. Implementation won’t be an issue because AI Rudder’s platform also has an open API that makes integration and deployment fast, easy and seamless.

Change is upon us. Are you ready?

With businesses overwhelmed by customer requests during Covid-19, Voice AI has proven to be a natural choice – especially in financial services and banking – to provide the agility needed to deal with the business impact of the pandemic.

AI Rudder’s enterprise-ready Voice AI assistants can help you face tomorrow’s challenges today. Visit us at www.airudder.com to find out more or contact us to arrange a product demo.

INDUSTRY VIEW FROM AI RUDDER

Using AI in agriculture could boost global food security – but we need to anticipate the risks

As the global population has expanded over time, agricultural modernisation has been humanity’s prevailing approach to staving off famine.

A variety of mechanical and chemical innovations delivered during the 1950s and 1960s represented the third agricultural revolution. The adoption of pesticides, fertilisers and high-yield crop breeds, among other measures, transformed agriculture and ensured a secure food supply for many millions of people over several decades.

Concurrently, modern agriculture has emerged as a culprit of global warming, responsible for one-third of greenhouse gas emissions, namely carbon dioxide and methane.

Meanwhile, inflation on the price of food is reaching an all-time high, while malnutrition is rising dramatically. Today, an estimated two billion people are afflicted by food insecurity (where having access to safe, sufficient and nutrient-rich food isn’t guaranteed). Some 690 million people are undernourished.

The third agricultural revolution may have run its course. And as we search for innovation to usher in a fourth agricultural revolution with urgency, all eyes are on artificial intelligence (AI).

AI, which has advanced rapidly over the past two decades, encompasses a broad range of technologies capable of performing human-like cognitive processes, such as reasoning. It’s trained to make these decisions based on information from vast amounts of data.

Using AI in agriculture

In assisting humans in fields and factories, AI may process, synthesise and analyse large amounts of data steadily and ceaselessly. It can outperform humans in detecting and diagnosing anomalies, such as plant diseases, and making predictions including about yield and weather.

Across several agricultural tasks, AI may relieve growers from labour entirely, automating tilling (preparing the soil), planting, fertilising, monitoring and harvesting.

Algorithms already regulate drip-irrigation grids, command fleets of topsoil-monitoring robots, and supervise weed-detecting rovers, self-driving tractors and combine harvesters. A fascination with the prospects of AI creates incentives to delegate it with further agency and autonomy.

This technology is hailed as the way to revolutionise agriculture. The World Economic Forum, an international nonprofit promoting public-private partnerships, has set AI and AI-powered agricultural robots (called “agbots”) at the forefront of the fourth agricultural revolution.

But in deploying AI swiftly and widely, we may increase agricultural productivity at the expense of safety. In our recent paper published in Nature Machine Intelligence, we have considered the risks that could come with rolling out these advanced and autonomous technologies in agriculture.

From hackers to accidents

First, given these technologies are connected to the internet, criminals may try to hack them.

Disrupting certain types of agbots would cause hefty damages. In the US alone, soil erosion costs US$44 billion (£33.6 billion) annually. This has been a growing driver of the demand for precision agriculture, including swarm robotics, that can help farms to manage and lessen its effects. But these swarms of topsoil-monitoring robots rely on interconnected computer networks and hence are vulnerable to cyber-sabotage and shutdown.

Similarly, tampering with weed-detecting rovers would let weeds loose at a considerable cost. We might also see interference with sprayers, autonomous drones or robotic harvesters, any of which could cripple cropping operations.

Beyond the farm gate, with increasing digitisation and automation, entire agrifood supply chains are susceptible to malicious cyber-attacks. At least 40 malware and ransomware attacks targeting food manufacturers, processors and packagers were registered in the US in 2021. The most notable was the US$11 million ransomware attack against the world’s largest meatpacker, JBS.

Then there are accidental risks. Before a rover is sent into the field, it’s instructed by its human operator to sense certain parameters and detect particular anomalies, such as plant pests. It disregards, whether by its own mechanical limitations or by command, all other factors.

The same applies to wireless sensor networks deployed in farms, designed to notice and act on particular parameters, for example, soil nitrogen content. By imprudent design, these autonomous systems might prioritise short-term crop productivity over long-term ecological integrity. To increase yields, they might apply excessive herbicides, pesticides and fertilisers to fields, which could have harmful effects on soil and waterways.

Rovers and sensor networks may also malfunction, as machines occasionally do, sending commands based on erroneous data to sprayers and agrochemical dispensers. And there’s the possibility we could see human error in programming the machines.

Safety over speed

Agriculture is too vital a domain for us to allow hasty deployment of potent but insufficiently supervised and often experimental technologies. If we do, the result may be that they intensify harvests but undermine ecosystems. As we emphasise in our paper, the most effective method to treat risks is prediction and prevention.

We should be careful in how we design AI for agricultural use and should involve experts from different fields in the process. For example, applied ecologists could advise on possible unintended environmental consequences of agricultural AI, such as nutrient exhaustion of topsoil, or excessive use of nitrogen and phosphorus fertilisers.

Also, hardware and software prototypes should be carefully tested in supervised environments (called “digital sandboxes”) before they are deployed more widely. In these spaces, ethical hackers, also known as white hackers, could look for vulnerabilities in safety and security.

This precautionary approach may slightly slow down the diffusion of AI. Yet it should ensure that those machines that graduate the sandbox are sufficiently sensitive, safe and secure. Half a billion farms, global food security and a fourth agricultural revolution hang in the balance.

Asaf Tzachor, Research Affiliate, Centre for the Study of Existential Risks, University of Cambridge

This article is republished from The Conversation under a Creative Commons license. Read the original article.