Predictive policing has exposed a new group of future criminals: MEPs.
A new testing systems has spotlighted five EU politicians as “at risk” of committing future crimes. Luckily for them, it’s not a tool that’s used by law enforcement, but one designed to highlight the dangers of such systems.
The project is the brainchild of Fair Trials, a criminal justice watchdog. The NGO is campaigning for a ban on predicting policing, which uses data analytics to forecast when and where crimes are likely to happen — and who may commit them.
Proponents argue that the approach can be more accurate, objective, and effective than traditional policing. But critics warn that it hardwires historic biases, disproportionately targets marginalised groups, amplifies structural discrimination, and infringes on civil rights.
“It might seem unbelievable that law enforcement and criminal justice authorities are making predictions about criminality based on people’s backgrounds, class, ethnicity and associations, but that is the reality of what is happening in the EU,” said Griff Ferris, Senior Legal and Policy Officer at Fair Trials.
Indeed, the technology is increasingly popular in Europe. In Italy, for instance, a tool known as Dalia has analysed ethnicity data to profile and predict future criminality. In the Netherlands, meanwhile, the so-called Top 600 list has been used to forecast which young people will commit high-impact crime. One in three people on the list – many of whom have reported being harassed by police – were found to be of Moroccan descent.
To illustrate the impacts, Fair Trials developed a mock assessment of future criminal behaviour.
Unlike many of the real systems used by the police, the analysis has been made entirely transparent. The test uses a questionnaire to profile each user. The more “Yes” answers they give, the higher their risk outcome. You can try it out for yourself here.
Politicians from the Socialists & Democrats, Renew, Greens/EFA, and the Left Group were invited to test the tool. After completing the quiz, MEPs Karen Melchior, Cornelia Ernst, Tiemo Wölken, Petar Vitanov, and Patrick Breyer were all identified as at “medium risk” of committing future crime.
“There should be no place in the EU for such systems — they are unreliable, biased, and unfair.
The gang will face no consequences for their potential offences. In real-life, however, such systems could put them on police databases and subject them to close monitoring, random questioning, or stop and search. Their risk scores may also be shared with schools, employers, immigration agencies, and child protection services. Algorithms have even led people to be jailed with scant evidence.
“I grew up in a low-income neighbourhood, in a poor Eastern European country, and the algorithm profiled me as a potential criminal,” Petar Vitanov, an MEP from the Bulgarian Socialist Party, said in a statement.
“There should be no place in the EU for such systems — they are unreliable, biased, and unfair.”
Fair Trials released the test results amid growing calls to outlaw predictive policing.
The topic has proven divisive in proposals for the AI Act, which is set to become the first-ever legal framework on artificial intelligence. Some lawmakers are pushing for a total ban on predictive policing, while others want to give leeway to law enforcement agencies.
Fair Trials has given supporters of the systems a new reason to reconsider their views: the tech can also target them.
Big business opportunities are brewing in the cosmos. Morgan Stanley predicts the space economy will grow from €355 billion in 2020 to over €1 trillion by 2030 — and competition for the rewards is fierce.
The USA remains a celestial superpower, while China is emerging as a powerful challenger. Europe has historically lagged behind the world leaders — but is now carving out a promising niche.
Across the continent, countries are converging around a single segment of the market: small satellites in low-Earth orbit (LEO).
The maximum altitude in LEO is about one-third of Earth’s radius. Credit: ESA
As the name suggests, low-Earth orbits are relatively close to the globe’s surface: a maximum of 2,000km above the planet, and sometimes as low as 160km. Commercial planes, by comparison, rarely fly at altitudes much higher than 14km.
In the 50 years since astronauts last stepped on the Moon, human space exploration has been confined to LEO. Crewless probes still fly deeper into our solar system, but most satellites — as well as the International Space Station — are now found in low-Earth orbit.
The LEO appeal
Small satellites in LEO may lack the glamour of spaceships taking astronauts to the moon, but they offer compelling advantages.
The lower altitude alone has numerous attractions. The costs, risks, and time required for more distant missions have reduced their allure, while the appeal of low-Earth orbit has increased. Among its advantages are speed boosts from gravity’s pull; better signal-to-noise ratios for radar and lidar; higher geospatial position accuracy; expanded launch vehicle options; more convenient journeys and — crucially — fewer resource needs.
‘The pandemic highlighted the need for high-speed connectivity.
Investments have surged as the use cases have expanded. LEO can provide internet connectivity, Earth observation, satellite navigation, and weather forecasting — and it’s becoming more accessible.
As a result, the number of projects in low-Earth Orbit is increasing rapidly. Dan York, who led the Internet Society’s 2022 LEO satellite report, attributes this growth to three key factors: the ceaseless demand for connectivity, the plummeting costs of satellites, and an expanding funding pool.
“The pandemic highlighted the need for high-speed connectivity that can be used for video communication, online learning, e-commerce, and more,” York told TNW. “LEO satellite systems have emerged as a powerful way to provide that high-speed, low-latency connection.”
In the race to commercialise LEO, a single target has been assigned pivotal significance: the first-ever orbital launch from Western Europe.
The territorial advantage
Europe already has a functioning equatorial spaceport — in South America. The Guiana Space Centre in Kourou, French Guiana, has been in operation since 1968. Originally, it served as the spaceport of France, but it’s now shared with the European Space Agency (ESA), which covers two-thirds of its budget.
Despite being 6,000km from mainland Europe, the site has a propitious location. Its position near the equator reduces the energy required for geostationary orbits, which match the rotation of the Earth. Rockets launching east can harness this momentum, while the centre’s proximity to open sea reduces risks to human habitations.
In Western Europe, however, a satellite has still never been sent into orbit — but the milestone is getting closer.
Polar orbits can offer comprehensive views of Earth.
The achievement would provide more mere than bragging rights. A homegrown spaceport would be a powerful launchpad for a budding LEO sector.
The location also has advantages. Western Europe can harness Earth’s rotation to power polar orbits, a flight path that passes the planet from north to south. This trajectory gives satellites extensive views of the planet rotating below, which is particularly useful for observation, mapping, and surveillance.
Further benefits would arise from the proximity to Europe’s production sites, talent, and connected industries.
“For the first time, the EU will have its own telecommunications constellation.
The war in Ukraine has exposed another lure of LEO. As a result of Russia’s full-scale invasion, Ukraine’s terrestrial internet connection has been disrupted by damage, outages, and jamming. In response, Elon Musk’s SpaceX offered free access to the Starlink satellite internet system, which has kept the country connected.
“The success of SpaceX’s Starlink service throughout Europe, and particularly in Ukraine, has shown the power of LEO satellite systems,” said York.
The EU is now developing its own satellite constellation. Known as IRIS2, the network is designed to maintain internet access during crisis situations. The $6.2 billion project is scheduled to launch by 2027.
“For the first time, the European Union will have its own telecommunications constellation, in particular in low orbits, the new frontier for telecommunication satellites,” said MEP Christophe Grudler, rapporteur on the EU secure connectivity programme.
The bloc has grand plans to compete with Starlink — and that’s just one of Europe’s LEO ambitions.
All around the continent, countries are trying to reap the benefits. The first one to reach orbit will get an edge over the competition.
Contenders in the race
Only nine countries and one international organisation (the aforementioned ESA) currently have orbital launch capability, according to the Pentagon.
Booming demand is expected for their services. The number of operational satellites is projected to grow from 5,000 today to 100,000 by 2040 — and spaceports across Europe are sprouting up to launch them.
Among them is a Spaceport Cornwall in the UK. In January, the site tried to send a satellite into orbit, but the attempt ended in bitter disappointment. After the Virgin Orbit rocket was successfully released, an engine malfunction brought the mission to a premature close.
The failure was a painful setback for Britain’s launch sector, but by no means a fatal one. Virgin Orbit is considering another go in Cornwall, while the SaxaVord spaceport in the Shetland Islands is set to attempt a launch before the end of the year. Further sites are under development in Sutherland, Argyll, Prestwick, Snowdonia, and the Outer Hebrides.
Virgin Orbit releases its rockets from a plane, in what’s known as a horizontal launch. Credit: Virgin Orbit
The UK does, however, face growing competition from spaceports in the EU. Most are located in isolated areas of Northern Europe, where populations are sparse and the sea is close.
Sweden’s Spaceport Estrange, for instance, recently became Europe’s first mainland satellite launch facility. The inaugural take-off from the complex is expected in late 2023.
“Europe has its foothold in space and will keep it,” said EU Commission President Ursula von der Leyen at the centre’s opening in January.
Another contender in the race is Andøya Space in Norway, which hopes to launch its first satellite rocket this year. Sites in Iceland, the Azores, Andalusia, the Canary Islands, and the North Sea are also in the running.
“This is exactly the infrastructure we need, not only to continue to innovate but also to explore the final frontier further.” -Ursula von der Leyen, President of the @EU_Commission
“We’re seeing a proliferation of space bases in Europe,” Marie-Anne Clair, head of the Guiana Space Centre, told AFP in December. “The commercial aspect is real: there is also an abundance of micro-satellites which will require missions from micro-launchers.”
An LEO island
A satellite launch provides a springboard for the host nation’s space sector — even if it fails. The UK’s attempt last month, for instance, added impetus to the local industry.
Before malfunctioning, the rocket did reach space, while Spaceport Cornwall became the world’s newest space launch operations centre. The build-up also boosted domestic satellite development and connected an array of talent, businesses, and public sector organisations.
Among them is Open Cosmos, an Oxfordshire-based startup that had a satellite onboard the Virgin Orbit rocket.
“We delivered a satellite in record time,” Open Cosmos CEO Rafel Jordá Siquier told TNW. “It’s sad not to have it in orbit, but we’re ready to come back and rebuild the satellite at an even faster pace if needed.”
Jordá founded Open Cosmos in 2015, with the aims of producing satellite-based solutions for Earth’s biggest problems.
Despite the ill-fated launch, Jordá believes that the UK is now a major player in LEO.
“The UK, with its leadership and its growing commitment towards the space sector, now has a seat at the global table in this industry. And it’s very important that we keep our team there and keep developing our capabilities.”
Those capabilities now encompass downstream applications, data and information services, and the upstream satellite and launch capabilities.
According to Jordá, the UK’s small satellite technology is particularly impressive. Alongside startups such as Open Cosmos, the country is home to OneWeb, one of the world’s leading satellite internet players. In January, the company announced that it now had 542 satellites in orbit – more than 80% of the fleet for its first-generation constellation.
“We now have a truly pan-UK capability.
The landscape also encompasses Surrey Satellite Technologies’ world-leading small satellite platforms, alongside first-rate “CubeSat” nanosatellites produced by Open Cosmos, Clyde Space, and Spire.
Another drawing card is a strong supply chain for space hardware and software. British luminaries in this area range from SMEs such as Teledyne, to aerospace giants like Airbus UK and BAE Systems.
Paul Febvre, CTO at Satellite Applications Catapult and a professor at Bradford University’s new space AI centre, said launch sites will complete the package.
“Now that we are establishing small-satellite launch facilities in both Cornwall and Scotland, with new ventures being developed in East Anglia, we have a truly pan-UK capability which creates the conditions for competition and success,” Febvre told TNW.
The new Bradford-Renduchintala Centre for Space AI plans to launch its own ‘pocket cube’ into LEO in 2024.
Until the British sites are operational, Open Cosmos will take off from other nations. But in the near future, Jordá plans further launches from the UK and France.
“The nice thing about the launch landscape at the moment is that it’s very diverse,” he said. “It’s getting very competitive and that means we have multiple partners that we can work with for different types of orbits.”
Space on the mainland
Another nation with eyes on LEO is France, which has the largest national space programme in Europe.
“France’s space startup ecosystem is particularly strong in LEO satellites, and I think we will see a number of winners emerging from there,” Maureen Haverty, VP at Seraphim, a prolific investor in space tech startups, told TNW.
“France is also the most successful country at encouraging US companies to set up in Europe and is the European hub for a number of key players.”
“We will have our own SpaceX.
A further asset for France is Arianespace, Europe’s leading satellite space launch company. The aerospace giant is currently developing a new reusable rocket, called Maïa, to challenge SpaceX. The launcher is due to be operational by 2026.
“For the first time Europe… will have access to a reusable launcher,” said French Finance Minister Bruno Le Maire last year. “In other words, we will have our SpaceX, we will have our Falcon 9.”
Critics have dismissed the prospect of Maïa competing with SpaceX, but the rocket would at least offer a European alternative. PLD Space in Spain plans to provide another.
The company aims to produce Spain’s first rocket to reach orbit — as well as Europe’s first reusable launch vehicle.
Named Miura 1, the 12.5-meter tall vehicle has a payload capacity of 100kg — just a fraction of the SpaceX Falcon 9’s 25,000kg and the Rocket Lab Electron’s 200–300kg. But PLD Space is confident it can serve the booming demand for small payload launches to LEO.
“We have demonstrated that PLD Space is the most promising company to improve European competitiveness in the microlaunchers race to space,” Ezequiel Sánchez, the company’s executive president, said after a successful test last year.
“This fact makes our project strategic not only for Spain but with a European perspective and a reference to show the profitability of reinforcing investment in new players.”
The public investment in Arianespace provides an edge over rivals that rely on private funding — but the competition is growing.
Smaller startups are also fighting for a spot in low-Earth orbit. Haverty, who oversees Seraphim’s investments in LEO satellite businesses, has seen success from those that tap into domestic expertise.
As an example, she points to Seraphim portfolio company ICEYE, which applies Finland’s strengths in hardware and avionics to world-leading synthetic aperture radar. In Lombardy, meanwhile, D-Orbit has harnessed Italy’s space heritage to establish the planet’s only commercial in-space delivery orbital tug company. The firm recently launched its seventh and eighth missions on SpaceX’s transporter mission.
Prior to joining Seraphim, Haverty was COO at Apollo Fusion, a space startup that was acquired in 2021 for $145M by Astra. She then worked on investments for Astra.
“Overall, I think Europe’s route to success lies in focusing on where they can be world leaders rather than trying to develop a European alternative to an American solution,” said Haverty.
Points of differentiation could also compensate for some shortcomings.
Rejuvenating the old world
The barriers to LEO are lowering, but they remain daunting. Funding in Europe still can’t compete with what’s in the US; space projects are susceptible to delays that can push customers to bigger competitors; nascent markets are tricky to target with commercially viable products.
Startups providing satellite internet services face further obstacles. The established leaders of SpaceX and OneWeb already have LEO constellations in space, customer equipment available, and regulatory approvals in many countries.
“The single biggest challenge for these [European] projects is to get all the components launched and in orbit,” said York, who led the Internet Society’s recent LEO report. “The second biggest challenge is to obtain the regulatory approvals in every country in which they want to operate.”
The growing demand for sparse skills is also difficult to meet. Rewards on offer in the US are often far more lucrative, and the finance and gaming sectors suck up much of the top tech talent.
“We need to encourage and stimulate the understanding in our new generation of engineers, innovators, and entrepreneurs that space is a fantastic place to develop a career and business opportunities, and universities are the knowledge engine for the future economy,” said Febvre, CTO at Satellite Applications Catapult.
At Satellite Applications Catapult, Febvre focuses on growing the UK space industry.
Further problems have emerged in Europe’s commercial launch sector. As well as lacking spaceports, the continent is short on effective rockets. Arianespace’s Vega launcher has been marred by repeated failures, the Araine-5 rocket will soon be retired, and its replacement may not be available for over a year.
The rocket shortage could delay the launch of satellites into LEO. Consequently, Europe has become reliant on commercial launch partners, particularly SpaceX.
“Europe does not have the ‘SpaceX Mafia’ effect.
Haverty adds two further weaknesses compared to the US: a limited product focus in satellite projects and a scarcity of second-generation spacetech founders.
“Europe does not have the ‘SpaceX Mafia’ effect,” she said. “European governments focus more on grants rather than on contracts, which makes it harder to grow startups into big businesses.”
The growing popularity of LEO has created another problem. There’s only so much space in space – and it’s starting to get crowded.
The ESA estimates that there are 36,500 chunks of space debris larger than 10cm, and 130 million between 1mm and 1cm. As these numbers grow, so do the risks of crashes and light pollution.
These threats can also scupper business plans. Regulators consider environmental concerns before deciding whether to allow a satellite launch, but their rules can place heavy burdens on LEO startups. Haverty hopes regulators exert greater pressure on the major operators than their smaller challengers.
“It’s important to remember that the vast majority of debris and causes of potential collisions are caused by the deliberate destruction of satellites by China and Russia,” she said. “Most operators are doing their best to keep space clean.”
On the plus side, the problems of space debris and pollution are presenting business opportunities for space robotics, manufacturing, and in-space servicing. European startups have pitched a range of solutions, from AI monitoring of debris to towing satellites out of LEO.
The ESA and Swiss startup ClearSpace plan to complete the first removal of space debris from orbit in 2025. Credit: ESA
Cleaning up is one of many emerging opportunities in LEO — and Europe is well-poised to grab a share
Despite the challenges, the continent has an enviable array of major satellite operators, affordable engineering talent, a rich history of multinational efforts, and a superlative satellite supply chain.
Insiders hope the spaceport race will further stimulate the sector. The first country across the line may not end up as the best, but the competition can be a boon for all contenders.
The capital flooding into LEO suggests the prospects are strong. Across the continent, investors are betting that a rising tide will lift all spaceships.
From steam power and electricity to computers and the internet, technological advancements have always disrupted labor markets, pushing out some careers while creating others. Artificial intelligence remains something of a misnomer — the smartest computer systems still don’t actually know anything — but the technology has reached an inflection point where it’s poised to affect new classes of jobs: artists and knowledge workers.
Specifically, the emergence of large language models – AI systems that are trained on vast amounts of text – means computers can now produce human-sounding written language and convert descriptive phrases into realistic images. The Conversation asked five artificial intelligence researchers to discuss how large language models are likely to affect artists and knowledge workers. And, as our experts noted, the technology is far from perfect, which raises a host of issues — from misinformation to plagiarism — that affect human workers.
To jump ahead to each response, here’s a list of each:
Lynne Parker, Associate Vice Chancellor, University of Tennessee
Large language models are making creativity and knowledge work accessible to all. Everyone with an internet connection can now use tools like ChatGPT or DALL-E 2 to express themselves and make sense of huge stores of information by, for example, producing text summaries.
These new AI tools can’t read minds, of course. A new, yet simpler, kind of human creativity is needed in the form of text prompts to get the results the human user is seeking. Through iterative prompting — an example of human-AI collaboration — the AI system generates successive rounds of outputs until the human writing the prompts is satisfied with the results. For example, the (human) winner of the recent Colorado State Fair competition in the digital artist category, who used an AI-powered tool, demonstrated creativity, but not of the sort that requires brushes and an eye for color and texture.
While there are significant benefits to opening the world of creativity and knowledge work to everyone, these new AI tools also have downsides. First, they could accelerate the loss of important human skills that will remain important in the coming years, especially writing skills. Educational institutes need to craft and enforce policies on allowable uses of large language models to ensure fair play and desirable learning outcomes.
According to McDermott,Concretene used in real construction projects was up to 30-50% stronger than standard concrete. Subsequent lab tests have shown strength gains that surpass 100%. As a result, the volume of cement required can significantly reduce without impairing performance.
Costs, shrinking, and cracking can also be trimmed, while increases in density cut concrete’s porosity.
“These factors will enable engineers to reduce the volume of concrete required in designs going forward, further reducing the CO₂ impact,” says McDermott.
Ultimately, he believes Concretene could be used in more than 99% of concretes worldwide.
McDermott’s team worked with the University of Manchester’s Graphene Engineering Innovation Centre (GEIC) to develop Concretene. Credit: Concretene
Initial tests of Concretene have produced promising results.
In total, more than 1,000 tonnes of Concretene have now been poured in real-world projects. The next target is pushing the product into the mainstream.
Graphene has been challenging to produce at speed and scale, but recent breakthroughs have made mass-production a more realistic prospect.
These developments have restored faith from investors. In December, Concretene raised £8m from a VC funding round led by LocalGlobe. The cash will be used to accelerate the commercialization plans.
Concretene will now turn to collaborations with Heathrow Airport, Network Rail, Yorkshire Water, and the Environment Agency. Work will also soon begin on the first road trial of Concretene with National Highways
“Concretene is being put through ‘live’ trials in the places where it can be used most,” says McDermott. “This is a key step to achieving industry adoption.”
That could be good news for both the construction sector and the environment. Because despite the claims of British shock jocks, you can’t actually grow concrete — but you cangrow graphene.
The UK’s space sector is searching for positives after the first orbital launch from western Europe ended in failure.
The mission appeared to have started smoothly. At around 10PM GMT on Monday, the Boeing 747 carrying Virgin Orbit’s LauncherOne rocket successfully took-off in southwest England.
The jet then climbed around 35,000ft before releasing the rocket over the Atlantic Ocean. But then, disaster struck.
“We appear to have an anomaly that has prevented us from reaching orbit. We are evaluating the information,” Virgin Orbit announced on Twitter.
We appear to have an anomaly that has prevented us from reaching orbit. We are evaluating the information.
The US company soon provided further details. The problem had emerged during the firing of LauncherOne’s second-stage engine, while the rocket was traveling at more than 11,000 mph.
All nine satellites onboard were lost. Among them was Amber-1, which was developed by the UK’s Satellite Applications Catapult and Horizon Technologies for maritime tracking.
“We will come back stronger.
Paul Febrve, CTO at Satellite Applications Catapult, said the failure was a big setback for everyone involved, but a “minor dent” to the UK’s space strategy.
“It’s a blow, but it’s not a crippling blow,” Febrve told TNW. “We will learn from it, come back stronger, and improve the capability that we’ve got in the UK.”
That capability has firm foundations. As an island with a northern latitude, Britain has auspicious geography for launching satellites into polarand sun-synchronous orbits, which go over the north and south poles.
There are several compelling reasons to harness these strengths. One is the growing demand for digital connectivity across the globe, which can’t be met by using terrestrial infrastructure alone.
Satellite Applications Catapult was set to operate two satellites on the British mission. Credit: Horizon Technologies
Another motive has been brutally highlighted by Russia’s war in Ukraine. The February invasion exposed the need to quickly deploy small satellites for military intelligence, which has increased demand for launches in Europe.
The UK has pitched itself as the ideal provider of these spaceports. In addition to its favorable geography, the country is a leading satellite producer, home to many private space companies, and the first nation in Europe to implement new spaceflight laws.
Seven spaceports across Britain are now under development. They’re unlikely to provide launchpads for missions to the moon, but they could offer promising locations for smaller satellites.
“This particular vehicle was carrying satellites from seven different providers, all doing different things. They were really handcrafted in terms of their purpose,” said Febvre.
“We’re really focused on responsive launch and innovation — not the scale-up aspect.”
“We remain committed to becoming Europe’s leading provider.
Febvre found further cause for hope is what’s already been achieved. While Monday’s mission didn’t achieve its ultimate goal, it did prove that space launches are achievable from British soil.
The attempt will also enhance domestic expertise, regulation, and capabilities.
“The project has succeeded in creating a horizontal launch capability at Spaceport Cornwall, and we remain committed to becoming the leading provider of commercial small satellite launch in Europe by 2030, with vertical launches planned from Scotland,” Matt Archer, director of commercial spaceflight at the UK Space Agency, said in a statement.
As the satellites were insured, their manufacturers and operators will be compensated for their loss. A bigger issue will be the reputational damage.
Setbacks in space are not unusual, but they still spook investors. Virgin Orbit now has to convince the critics that the failure won’t be repeated. The UK, meanwhile, is already planning another launch within the next 12 months.
We did it! Despite humanity’s best efforts, we made it through 2022. Before we pick ourselves up, dust ourselves off, and brace for whatever 2023 has to offer, we should probably take some time to reflect on the year that was.
Here at Neural, that means recounting our favorite stories from the past 12 months. There was a lot of mind-blowing news in the world of tech in 2022. From Elon Musk’s purchase of Twitter to former Google engineer Blake Lemoine declaring that he’d met a sentient AI, it was a year to remember.
But, rather than rehash months-old news, we wanted to take this opportunity to share our most mind-blowing and fascinating stories from the year. Some of these were big news when we published them, others have a more evergreen feel to them. But they’re all articles we’re particularly proud of.
Greetings, humanoids
Subscribe to our newsletter now for a weekly recap of our favorite AI stories in your inbox.
So, before we bid you and 2022 adieu, and without further ado, here Neural’s most mind-blowing stories of the year.
Thomas Macaulay
TNW/Neural reporter Thomas Macaulay had a fantastic year. It’d take a few volumes to reprint all the amazing stories he wrote in 2022, but we’ve managed to snag five hits for your reading pleasure.
Why have aliens never visited Earth? Scientists blame the singularity
What if the real reason why we haven’t met aliens yet is because their civilizations became too big to succeed? This is all theoretical, but you might be surprised who the first species to experience this “burnout” could be.
It’s all fun and games until your childhood pal becomes a killer robot. This is eerily similar to the plot of the 1986 cult classic “Deadly Friend.” Although the real story is about a guy who trained an AI-powered microwave to act like his imaginary friend, and the movie was about a guy who shoved a computer chip in a dead person’s head, both tales have their merits as classic horror features if you ask me.
This story is a delightful dive into what it’s like to experience culinary cuisine at the cutting edge. It’s a great story. But, if I’m being honest, my biggest takeaway is that Tom’s childhood favorite food was steak.
The Dutch are world leaders in lab-grown meat. Why can’t they eat it?
This is one of my favorites, and a perfect example of why Tom’s so good at what he does. This deep dive not only discusses the technology, but dares to ask hard questions: “It’s not vegetarian, but if it’s removed every drawback of conventional meat, why wouldn’t I eat it? And why can’t I find it in Europe?”
Elon Musk is the richest person in the world. If you ask us, that’s way less impressive than it sounds. He spent 2022 doing what he always does: making headlines and causing controversy. Rather than speculate about what he’s going to do next, we wanted to gently remind you that he has a habit of making things up as he goes along.
Tristan Greene
I’m not usually one to toot my own horn. But, since it’s the holiday season, I thought I’d share my favorite Neural stories that were written by yours truly in 2022:
This whole thing turned into a big deal on Twitter, at least as far as the AI community goes. There was a significant amount of respectful debate that has since boiled over into numerous other discussions about fancy AI models from OpenAI and Meta.
I’m not sure who needs to hear this but, your brain is way smarter than you think it is. While you’re enjoying life in a classical world, our brains are (theoretically) operating in a quantum one. If it sounds tricky, that’s because it is.
The biggest story of 2021 was Google’s time crystals. To date, I think it’s the most important story I’ve ever covered. But 2022 also had some very cool experiments in the same domain. I can’t wait for 2023!
Elon Musk and Tesla are trying to convince the world that they’re on the cusp of putting a humanoid helper robot into production. Spoiler alert: they most certainly are not. A little critical thinking goes a long way here.
I wrote this piece in early January of 2022 and having thought about it for the whole year, I stand by it. I’m pretty sure the world ended in 2012, it’s the only thing that makes any sense.
However, if it didn’t, and all of this has been real, then I’d like to wish you a wonderful 2023. On behalf of Neural, thanks for reading. Happy new year!
Here we go again! For the sixth year running, we present Neural’s annual AI predictions. 2022 was an incredible year for the fields of machine learning and artificial intelligence. From the AI developer who tried to convince the world that one of Google’s chatbots had become sentient to the recent launch of OpenAI’s ChatGPT, it’s been 12 months of non-stop drama and action. And we have every reason to believe that next year will be both bigger and weirder.
That’s why we reached out to three thought leaders whose companies are highly invested in artificial intelligence and the future. Without further ado, here are the predictions for AI in 2023:
First up, Alexander Hagerup, co-founder and CEO at Vic.ai, told us that we’d continue to see the “progression from humans using AI and ML software to augment their work, to humans relying on software to autonomously do the work for them.” According to him, this will have a lot to do with generative AI for creatives — we’re pretty sure he’s talking about the ChatGPTs and DALL-Es of the AI world — as well as “reliance on truly autonomous systems for finance and other back-office functions.”
Greetings, humanoids
Subscribe to our newsletter now for a weekly recap of our favorite AI stories in your inbox.
He believes a looming recession could increase this progress as much as two-fold, as businesses may be forced to find ways to cut back on labor costs.
Next, we heard from Jonathan Taylor, Chief Technology Officer at Zoovu. He’s predicting global disruption for the consumer buyer experience in 2023 thanks to “innovative zero-party solutions, leveraging advanced machine learning techniques and designed to interact directly and transparently with consumers.” I know that sounds like corporate jargon, but the fact of the matter is sometimes marketing-speak hits the nail on the head.
Consumers are sick and tired of the traditional business interaction experience. We’ve been on hold since we were old enough to pay bills. It’s a bold new world and the companies that know how to use machine learning to make us happy will be the cream that rises to the top in 2023 and beyond.
Jonathan Taylor, Chief Technology Officer at Zoovu
Taylor also predicts that Europe’s world-leading consumer protection and data privacy legislation will force companies large and small to “adopt these new approaches before the legacy approaches either become regulated out of existence by government or mandated out of existence by consumers.”
The writing’s on the wall. As he puts it, “the only way to make these zero-party solutions truly scalable and as effective as the older privacy-invading alternatives, will be to use advanced machine learning and transfer learning techniques.”
Finally, we got in touch with Gabriel Mecklenburg, co-founder at Hinge Health. He told us that the future of AI in 2023 is diversity. In order for the field to progress, especially when it comes to medicine, machine learning needs to work for everyone.
In his words, “AI is clearly the future of motion tracking for health and fitness, but it’s still extremely hard to do well. Many apps will work if you’re a white person with an average body and a late-model iPhone with a big screen. However, equitable access means that AI-powered care experiences must work on low-end phones, for people of all shapes and colors, and in real environments.”
Gabriel Mecklenburg, co-founder of Hinge Health
Mecklenburg explained that more than one in five people suffer from musculoskeletal conditions such as neck, back, and joint pain. According to him, “it is a global crisis with a severe human and economic toll.”
He believes that, with AI, medical professionals have what they need to help those people. “For example,” says Mecklenberg, “AI technology can now help identify and track many unique joints and reference points on the body using just the phone camera.”
But, as mentioned above, this only matters if these tools work for everyone. Per Mecklenburg, “we must ensure AI is used to bridge the care gap, not widen it.”
From the editor of Neural:
It’s been a privilege curating and publishing these predictions all these years. When we started, over half a decade ago, we made the conscious decision to highlight voices from smaller companies. And, as long-time readers might recall, I even ventured a few predictions myself back in 2019.
But, considering we spent all of 2020 in COVID lockdown, I’m reticent to tempt fate yet again. I won’t venture any predictions for AI in 2023 save one: the human spirit will endure.
When we started predicting the future of AI here at Neural, a certain portion of the population found it clever to tell creatives to “learn to code.” At the time, it seemed like journalists and artists were on the verge of being replaced by machines.
Yet, six years later, we still have journalists and artists. That’s the problem with humans: we’re never satisfied. Build an AI that understands us today, and it’ll be out of date tomorrow.
The future is all about finding ways to make AI work for us, not the other way around.
In the spring of 2010, physicist Jari Kinaret received an email from the European Commission. The EU’s executive arm was seeking pitches from scientists for ambitious new megaprojects. Known as flagships, the initiatives would focus on innovations that could transform Europe’s scientific and industrial landscape.
“I was not very impressed,” the 60-year-old tells TNW. “I thought they could find better ideas.”
Greetings, humanoids
Subscribe to our newsletter now for a weekly recap of our favorite AI stories in your inbox.
As it happened, Kinaret had an idea of his own: growing graphene. He decided to submit the topic for consideration.
That proposal lay the foundation for the Graphene Flagship: the largest-ever European research program. Launched in 2013 with a €1 billion budget, the project aimed to bring the “wonder material” into the mainstream within 10 years.
On the eve of that deadline, TNW spoke to Kinaret about the project’s progress over the past decade — and his hopes for the next one.
Graphene arrives in Europe
Scientists have pursued the single sheet of carbon atoms that constitute graphene since 1859, but its existence wasn’t confirmed until 2004. The big breakthrough was sparked by a strikingly simple product: sticky tape.
Andre Geim and Konstantin Novoselov, two physicists at the University of Manchester, would regularly hold “Friday night experiments,” where they’d explore outlandish ideas. At one such session, adhesive tape was used to extract tiny flakes from a lump of graphite. After repeatedly separating the thinnest fragments, they created flakes that were just one atom thick.
The researchers had isolated graphene — the first two-dimensional material ever discovered.
The researchers donated graphite, tape, and a graphene transistor to the Nobel Museum. Credit: Gabriel Hildebrand
The science world was abuzz with excitement. Graphene was the thinnest known material in the universe, the strongest ever measured, more pliable than rubber, and more conductive than copper.
In 2010, Geim and Novoselov won a Nobel Prize for their discovery. The award committee envisioned endless applications: touch screens, light panels, solar cells, satellites, meteorology, and, err, virtually invisible hammocks for cats.
The hypothetical hammock would weigh just 0.77 mg and support a 4 kg cat. Credit: Royal Swedish Academy of Sciences.
Kinaret recognized the potential. Three years later, he was heading an EU drive to take graphene from the lab to the market.
Hype versus reality
Commercializing graphene was never going to be straightforward. Studies suggest that innovations typically take between five and seven decades to evolve from inventions to products with significant market shares. Evolution would be slow — but observers were already impatient.
As the Flagship’s director, Kinaret had to manage such starry-eyed expectations. At talks, he’d frequently refer to the Gartner hype cycle, a depiction of how new technologies evolve.
The timeline starts with a breakthrough that sparks media excitement. In graphene’s case, reporters were soon claiming the material was set to replace silicon.
“Graphene cannot replace silicon,” says Kinaret. “Graphene is a semi-metal; it’s not a semiconductor.”
When reality fails to meet such inflated expectations, interest wanes and investment shrinks. Gartner describes this stage as the “trough of disillusionment.” Graphene appears to have exited this perilous period, partly thanks to the EU’s long-term commitment.
The backers that remain tend to be more practical and persistent. Now, their target is mainstream adoption.
“That’s something we promised — and delivered.
Initially, many commercial partners were frugal with their investments. One very large European company had a budget of only €20,000 for 30 months — “just enough to buy coffee for the people working on it, but not really enough to do anything substantial,” Kinaret recalls.
To increase their involvement, the Flagship needed their trust, which was challenging as rival brands would have to work together. Nokia, for instance, would have to collaborate with Ericsson.
“One dimension of trust that people needed was to trust this is for real,” says Kinaret. “The other is that participants needed to trust each other.”
The Flagship’s current membership suggests that trust has now been secured. The proportion of companies has grown from 15% to roughly 50% today. The other half are either research organizations or universities.
Kinaret describes the growth of industrial engagement as the Flagship’s key development.
“That’s something that we promised, and it’s something we have delivered,” he says.
From lab to fab
Around 100 products have emerged from the Graphene Flagship. The vast majority are business-to-business technologies, such as thermal coating for racing cars and eco-friendly packaging for electronic devices. Consumers’ products have been slower to commercialize.
Kinaret spotlights a few of his favorites. One is Qurv, a Spanish spinoff that makes graphene-based sensors, which cars can use to detect pedestrians in fog and rain.
“There are detectors today that do the same thing, but they can cost about $500 each,” says Kinaret. “The graphene detectors could cost about $1 each. That would be a total game changer in that business.”
Qurv’s wide-spectrum image sensors could enhance computer vision. Credit: The Graphene Flagship
Another highlight for Kinaret is Inbrain Neuroelectronics. The startup is developing graphene-based implants that can monitor brain signals and treat neurological disorders.
The devices could eventually stimulate the brain to control tremors caused by Parkinson’s disease. Traditional electrodes can achieve this, but they’re far stiffer than highly-flexible graphene.
“The brain is like a lump of jelly — it keeps moving around,” says Kinaret. “If you put a stiff electrode there, it results in scar formation.
Kinaret is also excited about the prospects for fundamental science. In 2018, Graphene Flagship partners revealed that over 2,000 materials can exist in a 2D form. Not all of them are stable, but a number of them are the focus of active research.
“You can make superconducting materials.
Some researchers are exploring what can be achieved by stacking the substances in multi-layers.
“You can grow them so there is a very specific twist angle between the different layers, which means they’re slightly misaligned. This misalignment angle is a very important new parameter,” says Kinaret.
“By tuning this misalignment angle, you can make materials that are superconducting and that have very interesting optical properties. This has only been explored for roughly four years, in terms of basic research, and it’s still quite far from applications. But it offers interesting possibilities for the future.”
Mission accomplished?
Kinaret is proud of the Flagship’s achievements. He believes the initiative has surpassed its targets by significant margins.
The data appears to support his claims. The European Commission aims to turn every €10 million that’s invested into one patent application. The Flagship, says Kinaret, has more than 10 times that requirement. The targets for scientific publications, he adds, have been surpassed by a similar factor.
Kinaret’s research targets potential applications. Credit: Graphene Flagship
There are still challenges to overcome. In electronics, for instance, high-quality graphene has to be transferred from the substrate on which it’s grown and onto the system where it’s used. The Flagship can do that well manually, but automating the process on an industrial scale has proven more difficult.
Nonetheless, Kinaret reminds the team they should remain positive.
“Engineers are typically short-term optimists and long-term pessimists,” he says. “They expect progress to be much faster initially than it turns out to be, but in the end, they underestimate the impacts of new technologies.”
In the future, Kinaret expects Europe to become a graphene powerhouse. The Flagship has given the continent a head start over the US in the race toward the mainstream.
He admits, however, that laypeople still ask what graphene is and can do.
“If we get to a situation where a surprised ‘what?’ has been replaced by ‘so what?’ because it’s become ubiquitous or mainstream… then we’ll have made it.”
The war in Ukraine has become the largest testing ground for artificial intelligence-powered autonomous and uncrewed vehicles in history. While the use of military robots is nothing new — World War II saw the birth of remote-controlled war machines and the US has deployed fully-autonomous assault drones as recently as 2020 — what we’re seeing in Ukraine is the proliferation of a new class of combat vehicle.
This article discusses the “killer robot” technology being used by both sides in Russia’s war in Ukraine. Our main takeaway is that the “killer” part of “killer robots” doesn’t apply here. Read on to find out why.
Uncrewed versus autonomous
This war represents the first usage of the modern class of uncrewed vehicles and automated weapons platforms in a protracted invasion involving forces with relatively similar tech. While Russia’s military appears, on paper, to be superior to Ukraine’s, the two sides have fielded forces with similar capabilities. Compared to forces Russia faced during its involvement in the Syrian civil war or, for example, those faced by the US during the Iraq and Afghanistan engagements, what’s happening on the ground in Ukraine right now demonstrates a more paralleled engagement theater.
Greetings, humanoids
Subscribe to our newsletter now for a weekly recap of our favorite AI stories in your inbox.
It’s important, however, to mention that this is not a war being fought by machines. It’s unlikely that autonomous or uncrewed weapons and vehicles will have much impact in the war, simply because they’re untested and, currently, unreliable.
Uncrewed vehicles and autonomous vehicles aren’t necessarily the same thing. While almost all autonomous vehicles — those which can operate without human intervention — are uncrewed, many uncrewed vehicles can only be operated remotely by humans. Perhaps most importantly, many of these vehicles have never been tested in combat. This means that they’re more likely to be used in “support” roles than as autonomous combat vehicles, even if that’s what they were designed to do.
But, before we get into the how’s and why’s behind the usage of military robots in modern warfare, we need to explain what kind of vehicles are currently in use. There are no “killer robots” in warfare. That’s a catch-all term used to describe military vehicles both autonomous and uncrewed.
These include uncrewed aerial vehicles (UAVs), uncrewed ground vehicles (UGVs), and uncrewed surface vehicles (USVs, another term for uncrewed maritime or water-based vehicles).
So, the first question we have to answer is: why not just turn the robots into killers and let them fight the war for us? You might be surprised to learn that the answer has very little to do with regulations or rules regarding the use of “killer robots.”
To put it simply: militaries have better things to do with their robots than just sending fire downrange. That doesn’t mean they won’t be tested that way, there’s already evidence that’s happened.
A British “Harrier” USV, credit: Wikicommons
However, we’ve seen all that before. The use of “killer robots” in warfare is old hat now. The US deployed drones in Iraq and Afghanistan and, as we reported here at TNW, it even sent a Predator drone to autonomously assassinate an Iranian general.
What’s different in this war is the proliferation of UAVs and UGVs in combat support roles. We’ve seen drones and autonomous land vehicles in war before, but never at this scale. Both forces are using uncrewed vehicles to perform tasks that, traditionally, either couldn’t be done or require extra humanpower. It does also bear mentioning that they’re using gear that’s relatively untested, which explains why we’re not seeing either country deploying these units enmasse.
A developmental crucible
Developing wartime technology is a tricky gambit. Despite the best assurances of the manufacturers, there’s simply no way to know what could possibly go wrong until a given tech sees actual field use.
In the Vietnam war, we saw a prime example of this paradigm in the debut of the M-16 rifle. It was supposed to replace the trusty old M-14. But, as the first soldiers to use the new weapon tragically found out, it wasn’t suitable for use in the jungle environment without modifications to its design and special training for the soldiers who’d use it. A lot of soldiers died as a result.
A US Marine cleaning their M16 during the US-Vietnam War, credit: Wikicommons
That’s one of the many reasons why a number of nations who’ve so far refused any direct involvement in the war are eager to send cutting-edge robots and weapons to the Ukrainian government in hopes of testing out their tech’s capabilities without risking their own soldiers’ skin.
TNW spoke with Alex Stronell, a Land Platforms Analyst and UGV lead at Janes, the defense intelligence provider. They explained that one of the more interesting things to note about the use of UGVs, in particular, in the war in Ukraine, is the absence of certain designs we might have otherwise expected.
“For example, an awful lot of attention has been paid inside and outside of Russia to the Uran-9 … It certainly looks like a menacing vehicle, and it has been touted as the world’s most advanced combat UGV,” Stronell told us, before adding “however, I have not seen any evidence that the Russians have used the Uran-9 in Ukraine, and this could be because it still requires further development.”
On the other side, Stronell previously wrote that Ukrainian forces will soon wield the world’s largest complement of THeMIS UGVs (see the video below). That’s exceptional when you consider that the nation’s arsenal is mostly lend-leased from other countries.
Milrem, the company that makes the THeMIS UGV, recently announced that the German Ministry of Defence ordered 14 of its vehicles to be sent to the Ukrainian forces for immediate use. According to Stronell, these vehicles will not be armed. They’re equipped for casualty evacuation, and for finding and removing landmines and similar devices.
But it’s also safe to say that the troops on the ground will find other uses for them. As anyone who’s ever deployed to a combat zone can tell you, space is at a premium and there’s no point in bringing more than you can carry.
The THeMIS, however, is outfitted with Milrem’s “Intelligence Function Kit,” which includes the “follow me” ability. This means that it would make for an excellent battle mule to haul ammo and other gear. And there’s certainly nothing stopping anyone from rekitting the THeMIS with combat modules or simply strapping a homemade autonomous weapon system to the top of it.
As much as the world fears the dawning of the age of killer robots in warfare, the current technology just simply isn’t there yet. Stronell waved off the idea that a dozen or so UGVs could, for example, be outfitted as killer guard robots that could be deployed in the defense of strategic points. Instead, he described a hybrid human/machine paradigm referred to as “manned-unmanned teaming, or M-UMT,” where-in, as described above, unmounted infantry address the battlefield with machine support.
In the time since the M-16 was mass-adopted during an ongoing conflict, the world’s militaries have refined the methodology they use to deploy new technologies. Currently, the war in Ukraine is teaching us that autonomous vehicles are useful in support roles.
The simple fact of the matter is that we’re already exceptionally good at killing each other when it comes to war. And it’s still cheaper to train a human to do everything a soldier needs to do than it is to build massive weapons platforms for every bullet we want to send downrange. The actual military need for “killer robots” is likely much lower than the average civilian might expect.
However, AI’s gifts when it comes to finding needles in haystacks, for example, make it the perfect recon unit, but soldiers have to do a lot more than just identify the enemy and pull a trigger.
However, that’s something that will surely change as AI technology matures. Which is why, Stronell told us, other European countries are either currently in the process of adopting autonomous weaponry or already have.
In the Netherlands, for example, the Royal Army has engaged in training ops in Lithuania to test their own complement of THeMIS units in what they’re referring to as a “pseudo-operational” theater. Due to the closeness of the war in Ukraine and its ongoing nature, nearby nations are able to run analogous military training operations based on up-to-the-minute intel of the ongoing conflict. In essence, the rest of Europe’s watching what Ukraine and Russia do with their robots and simulating the war at home.
Soldiers in the Netherlands Royal Army in front of a Netherlands Royal Air Force AH-64 Apache helicopter, credit: Wikicommons
This represents an intel bonanza for the related technologies and there’s no telling how much this period of warfare will advance things. We could see innumerable breakthroughs in both military and civilian artificial intelligence technology as the lessons learned from this war begin to filter out.
To illustrate this point, it bears mention that Russia’s put out a one million ruble bounty (about €15,000) to anyone who captures a Milrem THeMIS unit from the battlefield in Ukraine. These types of bounties aren’t exactly unusual during war times, but the fact that this particular one was so publicized is a testament to how desperate Russia is to get its hands on the technology.
An eye toward the future
It’s clear that not only is the war in Ukraine not a place where we’ll see “killer robots” deployed enmasse to overwhelm their fragile, human, enemy soldier counterparts, but that such a scenario is highly unlikely in any form of modern warfare.
However, when it comes to augmenting our current forces with UGVs or replacing crewed aerial and surface recon vehicles with robots, military leaders are excited about AI’s potential usefulness. And what we’re seeing right now in the war in Ukraine is the most likely path forward for the technology.
That’s not to say that the world shouldn’t be worried about killer robots or their development and proliferation through wartime usage. We absolutely should be worried, because Russia’s war in Ukraine has almost certainly lowered the world’s inhibitions surrounding the development of autonomous weapons.
Thanks to the convergence of several trends and changes across different markets and industries, automation is becoming a critical factor in the success of businesses and products. Advances in artificial intelligence, in parallel with the accelerating digitization of all aspects of business, are creating plenty of opportunities to automate operations, reduce waste, and increase efficiency.
From managing your Information Technology (IT) bill to finding bottlenecks in your business processes and taking control of your own network operations, here are three areas where companies can gain from applying automation.
1. IT automation
Greetings, humanoids
Subscribe to our newsletter now for a weekly recap of our favorite AI stories in your inbox.
Practically every large organization has IT. Even small companies that don’t have in-house IT staff may pay for another company to do it for them. The growing demand for IT can put extra strain on professionals who must deal with the ever-expanding and changing landscape of application and compute platforms.
“I’ve never met an IT person or CIO who said they have so much time and budget that they can do everything the business asks and more. There’s always a shortage of ability to drive projects through IT,” says Bill Lobig, Vice President of IBM Automation Product Management.
The talent shortage is highlighting the need to provide automation tools to IT staff so they can manage application uptime and keep IT operations stable.
Fortunately, advances in artificial intelligence are helping companies move toward smart automation by gathering and processing all sorts of structured and unstructured data.
“We’re seeing companies have more confidence in applying AI to a broader set of data, including log files and metrics and information that are spinning off of the systems that are running in your business (databases, app servers, Kubernetes, VMs),” Lobig says.
Previously, IT experts may have optimized their infrastructure through informed judgments and overprovisioning their resources. Now, they can take the guesswork out of their decisions by using AI to analyze the data of the IT infrastructure, find patterns, estimate usage, and optimize their resources.
For example, J.B. Hunt, a logistics and transportation company, uses IBM Turbonomic software to automate the scaling of its cloud and on-premise resources. For their on-premises environment, J.B. Hunt is automating all non-disruptive actions 24×7 and scaling non-production actions during a nightly maintenance window.
“Workloads scale and spike—it’s not static. No matter how much performance testing and capacity you put into sizing an application deployment, it’s a guess, albeit an educated one. You don’t really know how your customers’ workloads are going to vary across different times,” Lobig says.
In their public cloud environment, the J.B. Hunt team has been using a combination of recommendations and automated actions to manage their resources. Over the course of 12 months, Turbonomic executed nearly 2,000 resizing actions which—assuming manual intervention requires 20 minutes per action—freed up over 650 hours of the team’s time to focus on strategic initiatives.
2. Business processes
Business processes are another area that can gain from advances in AI and automation. The previous wave of automation in business processes was mostly driven by robotic process automation (RPA). While RPA has had a tremendous effect on productivity, like other solutions, it has limits too.
RPA only addresses tasks that you think need automation. It can automate a poorly designed process but can’t optimize it. It also can’t handle tasks that can’t be defined through deterministic rules. This is where “process and task mining” enter the picture. According to Lobig:
RPA executes scripts to automate what you tell it to do. It’s very deterministic and rigid in what it can do, automating highly repeatable tasks. Process and task mining find inefficiencies you can’t see.
Process and task mining can answer questions such as, is your business really running the way you think it is? Is everyone completing processes in the same way? What should you optimize first? It helps you get past the low-hanging fruit and find the hidden inefficiencies of your business that can also be addressed with automation.
3. Networking
In the past, networking was a specialized hardware-based discipline largely controlled by big telecommunications companies. Today, the networking ecosystem is more complex as enterprises now require ubiquitous application distribution in a hybrid multi-cloud environment, from customer prem, to edge, to private and public clouds.
The challenge is deploying and connecting all application endpoints at scale. Networks must be agile and dynamic to maintain application performance, availability, security and user experience. Today’s networks, however, face unprecedented challenges that can render them unresponsive and unadaptable to change. Enterprise and service providers can address those needs, delivering custom enterprise network value with self-service enterprise control.
Organizations can now own and manage their networking functions and end-to-end connectivity without being experts in switches, routers, radio-access networks, and other hardware.
“Networking has become just another part of the application supply chain (like databases, VMs, and containers) that companies are already running. Why not have your network be part of your full IT landscape so that you can apply AI to optimize it?” Lobig says.
For example, consider a large multinational bank that provides its customers access to their accounts overseas through ATM machines. The company previously outsourced network connectivity to a big telco. When the telco faced an outage in one country where the bank provided service, the customers could not access their funds. Although the bank didn’t have control over the networking service, it was fined for the outage.
Now, thanks to software-defined wide area network (SD-WAN) and automation and orchestration tools such as IBM’s AIOps solutions and IBM SevOne Network Performance Management, the bank can assume control of its own software-defined network, instead of shifting such an important responsibility to another company. New application-centric network connectivity can enhance those capabilities. This can drive enhanced security, intelligent observability, and service assurance, while providing a common way to manage networks across the diversity of infrastructure, tools, and security constructs.
Another area of networking that will provide new opportunities for automation is 5G.
“A lot of people think about 5G as a faster networking technology. But 5G is going to transform and disrupt B2B use cases. It can really bring edge computing to the forefront,” Lobig says.
There’s an opportunity for organizations to leverage software-defined networking and 5G to unlock new business models where high-bandwidth, low latency, and local connectivity is crucial.
An example is DISH Wireless, a company that’s working with IBM to automate the first greenfield cloud-native 5G network in the US. DISH Wireless is using IBM’s network orchestration software and services to bring 5G network orchestration to its business and operations platforms. One application they’re working on is enabling logistics companies to track package locations down to the centimeter, thanks to edge connectivity, RFID tags, and network management software.
“We’re helping them do this with our telco and network computing automation, edge computing automation, and enabling them to set up state-of-the-art orchestration for their customers. These unexpected industries can use 5G to really transform how business gets done across different areas,” Lobig says.
Where is the industry headed?
Automation is quickly evolving and we’re bound to see many new applications in the coming months and years. For companies that are at the beginning of their automation journey, Lobig has a few tips.
In the business automation space, look at process and task mining. Do you really know where the time is being spent in your enterprise? Do you know how work is getting done? If you use this technology, you’ll be able to identify the patterns and sequence of events that go into good outcomes and those that go into bad outcomes. Armed with these insights, you can redesign and automate the processes that have the biggest impact upon your business.
Lobig also believes that IT automation will be a bigger theme in 2023 as the world faces an energy crisis and electricity costs potentially become an escalating problem. IT automation can help organizations to use the capacity they need, which may translate into savings.
IT automation can also be important in tackling the climate change crisis.
“These days, you can tell whether your organization’s data center or workload is running on a renewable energy source,” Lobig says. “With that data, IT automation has the potential to automatically move workloads from cloud to on-prem and back and across hyper-scalers to optimize for costs and efficiencies.”
As for the future, Lobig believes that low-code/no-code application platforms will play an important role in automation by enabling more employees to build the automations that can enhance productivity.
