Category: Technology

From a record-breaking AI startup in Paris to Croatian drones and Lithuanian food tech, Europe’s startup ecosystem has had a busy week.

The week of March 9th to 15th was by all accounts an exceptional week for European venture capital. Two deals alone, one in London and one in Paris, amounted to almost three billion dollars.

But beyond the headlines, the week really showed what European investor confidence looks like right now: AI infrastructure, cybersecurity, healthtech, defense and cross-border trade have all completed significant rounds.

The geography stretched from Vilnius to Zagreb to the Swiss Alps. However, the themes were unmistakably shaped by this moment.

The 💜 of EU technology

The latest rumors from the EU tech scene, a story from our wise founder Boris and questionable AI art. It’s free in your inbox every week. Register now!

Here are the ten most significant funding rounds in Europe this week.

1. Nscale – 1.7 billion euros Series C | London, UK

Start with the number: 1.7 billion euros or $2 billion. Not only is this the largest European funding round of the week, it is also the largest equity round ever raised by a European startup, according to the company.

The list of investors alone illustrates the extent of the ambitions: Nvidia, Citadel, Dell, Nokia, Jane Street and Point72 as well as the main financiers Aker ASA and 8090 Industries.

Nscale was founded a year ago in 2024 to underline how quickly development has evolved. Nscale builds a vertically integrated AI infrastructure, from GPU computing and networking to data services and orchestration software.

The Series C values ​​the company at $14.6 billion, a more than four-fold increase from its Series B valuation of $3.1 billion in September 2025. This September round itself has been described as record-breaking.

This round follows a €1.1 billion credit facility that Nscale signed in February. The company is building at a speed rarely seen before in the European tech ecosystem.

2. AMI Labs – Seed Funding Worth $1.03 Billion | Paris, France

The largest seed round ever raised by a European company and possibly anywhere else. Advanced Machine Intelligence, AMI, pronounced the French word for “friend,” announced its $1.03 billion capital raise on March 10, just four months after its founding.

The company’s chairman is Yann LeCun, the Turing Award-winning computer scientist who spent 12 years at Meta before leaving the company in November 2025 to build something he believes the entire AI industry is not ready to build.

The founding team includes former meta-AI researchers Saining Xie, Pascale Fung and Michael Rabbat. Strategic investors include Nvidia, Samsung, Toyota Ventures, Jeff Bezos and former Google CEO Eric Schmidt.

AMI has no product and no revenue. LeCun said the first year will be dedicated entirely to research. That investors transferred a billion dollars on these terms reflects both the credibility of the team and the sheer amount of capital now awaiting the next paradigm shift in AI.

3. Isembard – £37.5m Series A | London, UK

Less than a year after its seed round, London-based Isembard raised 37.5 million pounds ($50 million) to expand its network of AI-powered factories focused on high-precision manufacturing for the aerospace and defense industries.

The round was led by Union Square Ventures, with participation from Tamarack Global, IQ Capital, Notion Capital and CIV. Angel investors included Deel founder Alex Bouaziz and former Wise CFO Matt Briers.

Isembard’s model is unusual in manufacturing: The company owns and operates its own facilities and also operates a network of franchise locations based on MasonOS, its proprietary agent operating system for factory management.

4. Waiver – $33 million | Paris, France

Waiv is a spinout from Owkin, the French-American biotech company that launched as an independent entity this week in a $33 million funding round.

The company, previously known as Owkin Dx, develops AI-powered precision testing tools for oncology: It analyzes routine pathology slides and multimodal patient data to identify biomarkers and predict response to treatment.

What sets Waiv apart from the broader AI healthcare cluster is its focus on making tests already in the clinical workflow and routine slide analysis significantly more meaningful. Its products include RlapsRisk BC (breast cancer recurrence prediction), MSIntuit and BRCAura.

The company has existing partnerships with pharmaceutical companies and counts leading hospital systems among its customers. The spin-off as an independent entity is intended to accelerate commercial development without being constrained by a parent company’s strategic priorities.

5. Qevlar AI – $30M Series A | Paris, France

Security centers face a problem that is now well understood: too many alerts, too few analysts, too much time per investigation.

On average, manually investigating an alert in a large organization takes between 32 and 61 minutes. Qevlar AI claims its platform does this in under three minutes. On March 10, investors decided it was worth $30 million.

Founded in 2022 by Ahmed Achchak, the Paris-based startup has developed an agentic AI platform that connects to existing security tools (SIEM, EDR, CTI) and automates the entire investigation workflow at Tier 2 and Tier 3 depth. Instead of simply triaging alerts, the system creates a diagram-based understanding of the attack surface.

The round was co-led by Partech and Forgepoint Capital International, with EQT Ventures also participating. Forgepoint had led the company’s previous capital raise of $14 million, a sign of continued conviction.

The new capital will fund geographic expansion in EMEA and Asia Pacific as well as product development for predictive threat hunting.

6. Saltz – 20 million euros Series A | Vilnius, Lithuania

The food distribution sector in Europe is fragmented, largely offline and resistant to modernization. Founded in 2022 by Oberlo and Shopify veterans Andrius Šlimas, Tomas Šlimas and Reinis Štrodahs, Saltz seeks to do for professional kitchens what those platforms did for e-commerce retailers. On March 9, the company secured €20 million in Series A funding.

The platform connects restaurants and professional kitchens with food suppliers and bundles catalogs, orders, payments and logistics in a single interface. The company operates in approximately 20 countries and its customers include Hilton, Marriott International and independent restaurant operators.

The company plans to hire more than 100 employees across engineering, product, sales and operations by the end of 2026. It targets fresh and frozen foods, particularly meat and seafood, where supply chains are most complex and have the greatest scope for improvement.

7. Outpost – $17.5M Series A | London, UK

Cross-border selling has always been attractive in theory, but complicated in practice: VAT registrations, payment infrastructure that is not mobile, and tax liability in jurisdictions a merchant has never visited.

Outpost, a London-based platform founded by former Revolut executives, raised $17.5 million on March 10 to solve these problems at the infrastructure level.

The round was led by Ribbit Capital, the venture firm behind Revolut, Coinbase and Stripe. Outpost’s platform handles payments and tax compliance for merchants selling internationally, creating local legal entities and payment rails in the markets they enter so that the merchant has no direct liability.

This is where the context matters: Trade tariffs in 2026 have made cross-border trade simultaneously more attractive and legally more treacherous. Outpost builds on exactly this tension.

8. Orqa – €12.7 million Series A | Osijek, Croatia

The Croatian drone manufacturer Orqa has been building first-person view drone systems since 2018, initially for the enthusiast market and increasingly for defense.

On March 10, the company raised €12.7 million in a Series A led by Expeditions, the early-stage investor specializing in European security, with participation from Lightspeed Venture Partners, Taiwania Capital, Aymo and Radius Capital.

What sets Orqa apart in an increasingly crowded drone landscape is its level of vertical integration: The company designs and manufactures its own flight controllers, radios, motors, cameras and circuit boards, without components made in China. The Osijek factory currently produces up to 280,000 NDAA-compliant drones annually.

The Pentagon’s Drone Dominance Program plans to procure up to 300,000 small attack drones by 2027. Orqa CEO Srdjan Kovacevic has made it clear that the company is positioning itself to compete for these contracts.

9. Seprify – 13.4 million euros Series A | Freiburg, Switzerland

Seprify develops high-performance cellulose-based ingredients for industrial applications, targeting markets where synthetic materials face regulatory pressure or sustainability scrutiny.

The €13.4 million Series A, which closed this week, counts among its backers Inter IKEA Group, a well-known strategic investor in a materials company working on sustainable alternatives to petroleum-derived raw materials.

The Swiss deep tech sector has spawned a steady stream of university spin-offs in materials science, and Seprify fits this pattern: founded with roots in academic research, now on its way to commercial scale. The round will fund production capacity expansion and customer development.

10. Lemrock – seeds worth 6 million euros | Paris, France

If Nscale and AMI represent the biggest bets of the week, Lemrock is one of the most interesting. Founded in 2025, the Paris-based startup is building an infrastructure that enables brands to sell directly into conversational AI environments, ChatGPT, Claude, Perplexity and their equivalents.

On March 11, the company announced a €6 million seed round led by Galion.exe, with Criteo founder Jean-Baptiste Rudelle joining the board.

The company already works with more than 60 brands in Europe and the United States, including Maisons du Monde, Cdiscount and Engie, processing over 100 million interactions monthly.

The round is small compared to the others on this list. But the question Lemrock answers: What happens to retail when AI agents become the primary product discovery interface is a large one and one that has rarely been asked.

The team at the Chair of Learning Systems and Robotics at TUM (Technical University of Munich) has developed a modest-looking robot that resembles a stick on wheels with a camera at the top. However, don’t let the looks fool you. It could be one of the most useful robots designed for everyday people.

Led by Prof. Angela Schoellig, the team has built a robot that can find lost objects by creating and analyzing a spatial map of its surroundings. Next time you can’t find your keys or your glasses, don’t lose your mind because this robot will find them for you.

How does the robot find things?

The camera provides two-dimensional images, but these pixels also contain depth information. From this, the robot creates a centimeter-precise 3D map of its surroundings and constantly updates it as things change.

Technical University of Munich

A challenge with this approach is that objects are constantly being moved or replaced, causing the map to quickly become outdated. This causes the robot to scan the entire area again. To solve this problem, the researchers used an LLM-powered model to not only map the environment but also maintain and update the data.

It tracks objects and assigns a relevant rating. The score, time since the object was last seen, and other data points are then used to build a probabilistic model and decide which areas to scan and maintain.

Technical University of Munich

What makes it really clever is the layer of internet knowledge embedded in it. The robot understands that glasses are likely to be left on the table or windowsill, not on the stove or in the sink.

A language model then translates these real-world considerations into search probabilities, helping the robot focus on areas where the missing object is most likely to be located. This means the robot searches almost 30% more efficiently than when randomly scanning rooms.

What awaits the future of this robot?

Currently the robot is limited to open areas. The next challenge the team faces is teaching him to open drawers and cupboards so he can search in enclosed spaces.

It’s still early days, but a robot that truly understands your home and helps you find things in it feels more useful as a home robot than other AI robot projects we’ve seen in the past.

Two years after a brutal restructuring disempowered Motional’s workforce and shut down operations, the Hyundai-backed AV company is back on the Strip, still having a security operator for now but promising to remove one by the end of 2026.

Uber and Motional have relaunched a commercial robotaxi service in Las Vegas, making all-electric Motional IONIQ 5 vehicles available to riders at key locations on and around the Strip starting March 13, 2026.

The service marks a significant milestone for Motional, which completely shut down its commercial operations two years ago, cut about 40% of its workforce and was left struggling to survive after co-founder Aptiv withdrew its funding.

The relaunch is not yet completely driverless. Initially, Motional’s IONIQ 5 robotaxis will carry a human driver who will monitor the road from the driver’s seat.

The company expects to phase out the security operator and introduce a fully driverless service by the end of 2026. This will achieve the goal it set for itself during the restructuring in 2024.

TNW City Coworking Space – Where your best work happens

A workspace designed for growth, collaboration and endless networking opportunities at the heart of technology.

This is how the service works

Drivers requesting an UberX, Uber Electric, Uber Comfort or Uber Comfort Electric can be matched with a Motional IONIQ 5 at no additional cost. If there is a match, a notification appears in the app, giving drivers the option to accept the autonomous vehicle or switch to a conventional journey.

Users who want to maximize their chances of getting an AV can log in via the “Riding Preferences” section in their Uber app’s settings.

Once a robotaxi arrives, the vehicle can be unlocked and the ride started entirely via the Uber app.

Inside, audible warnings prompt drivers to close the doors and fasten their seatbelts. If assistance is needed at any time, a human assistance team is available via the Uber app.

At launch, the service will cover designated rideshare zones along Las Vegas Boulevard at Resorts World Las Vegas and Encore at the Wynn, as well as the Westgate Las Vegas Resort & Casino, curbside locations in downtown Las Vegas and the Town Square shopping district near the airport.

Both companies said they plan to expand their operating area, but did not provide further details.

The vehicle: SAE Level 4, FMVSS certified

The Robotaxi IONIQ 5 was jointly developed by Motional and the Hyundai Motor Group and is specifically designed for ride-hail applications. Uber says it is one of the first SAE Level 4-capable autonomous vehicles to be certified under the US Federal Motor Vehicle Safety Standards (FMVSS), the federal regulatory framework for motor vehicle equipment.

SAE Level 4 means the vehicle can handle all driving functions within a defined operational design area without human intervention, although it does not require the ability to operate everywhere under all conditions.

“This milestone reflects our shared commitment to introducing autonomous vehicles in a way that prioritizes safety, increases reliability and expands access to more driving options for our customers.” said Sarfraz Maredia, president of autonomous mobility and delivery at Uber

Motional’s way back: from near collapse to restart

The relaunch is the culmination of a turbulent two-year recovery. Motional was founded in 2020 as a $4 billion equal joint venture between Hyundai Motor Group and automotive technology company Aptiv.

The company conducted pilot rides in Las Vegas through Uber and Lyft and deliveries in Los Angeles through Uber Eats, all with a human security guard, and collected more than 130,000 autonomous rides through these programs.

The company’s problems crystallized in early 2024 when Aptiv announced it would stop allocating capital to the company, citing the high cost of commercializing robotaxi technology and the uncertain path to profitability. Aptiv had forecast a non-cash equity loss of around $340 million for 2024 alone.

With Aptiv’s withdrawal threatening to destabilize the entire company, Hyundai stepped in with a nearly $1 billion commitment: $475 million invested directly in Motional and $448 million to buy out 11% of Aptiv’s common equity stake. As a result of the restructuring, Hyundai owns approximately 85% of Motional’s common stock and Aptiv owns 15%.

The financing came with painful conditions. Motional halted all commercial rides and deliveries, paused plans to launch its second-generation driverless service and laid off about 550 employees, about 40% of its total workforce, across teams in Las Vegas, Pittsburgh, California and Massachusetts. T

The company focused solely on improving the underlying autonomous technology, including a shift to a more neural network-based approach to autonomy, before attempting a new commercial implementation.

Motional returned to fundraising in August 2025 with a $550 million Series B round led by Aptiv and backed by Hyundai and Nuance Investments, increasing the company’s valuation to $6.5 billion. This capital, together with the technological restructuring, forms the basis for today’s relaunch.

A busy week for Uber’s autonomous ambitions

The launch in Las Vegas is not a standalone announcement. That same week, Uber confirmed a deal with Zoox, Amazon’s autonomous vehicle subsidiary, to deploy Zoox’s purpose-built robotaxis on the Uber platform, initially in Las Vegas starting in summer 2026, followed by Los Angeles in mid-2027.

Uber and Wayve also announced a collaboration with Nissan on a robotaxi pilot in Tokyo, scheduled for late 2026. This would be Uber’s first autonomous vehicle partnership in Japan.

Uber says it is currently working with more than 25 autonomous vehicle partners in mobility, delivery and freight. The company announced in early 2026 that it plans to invest more than $100 million in charging infrastructure for autonomous vehicles.

The autonomous solutions division, founded in February 2026 under Maredia’s leadership, is focused on helping AV technology companies commercialize their deployments faster by providing demand generation, driver experience, customer support and fleet management services.

For Motional, the Las Vegas service is both proof and pressure test. The company’s technology, which has been quietly rebuilt and retrained since 2024, is now poised for its first sustained commercial deployment in the real world with paying drivers.

Amazon is reportedly planning to move its annual Prime Day shopping event earlier into the summer, moving the big sales event from its traditional July timing to the end of June. The change, first reported by Bloomberg, would mark one of the biggest timing adjustments to the company’s flagship retail promotion in years.

Amazon

Prime Day typically takes place in mid-July and has become one of the most important sales events of the year for Amazon and its marketplace sellers. The promotion was originally launched in 2015 to celebrate Amazon’s anniversary while encouraging more users to sign up for the company’s Prime subscription service.

Why should Amazon move Prime Day earlier?

Although Amazon hasn’t officially confirmed the postponement, the report suggests that the company wants to bring the event forward to the end of June, potentially reshaping the summer sales calendar for retailers.

Amazon

Moving Prime Day earlier could help Amazon stay ahead of competing summer promotions from retailers like Walmart, Target and Best Buy, which often run similar sales around the same time. In recent years, the event has also expanded significantly, sometimes expanding beyond the original two-day format to include multi-day sales and sparking competing promotions across the e-commerce industry.

What could this mean for buyers and sellers?

Prime Day has become one of Amazon’s biggest revenue-generating events, generating billions of dollars in sales while attracting new Prime members. Analysts have previously forecast that sales during the event window could generate tens of billions of dollars in global online spending.

Julie Clopper/Shutterstock

If the timing changes, it could shift the timing of when brands plan their summer promotions and when shoppers expect big discounts. For consumers, the biggest difference may simply be that Prime Day deals are arriving a few weeks earlier than usual. But for retailers that compete with Amazon, the impact could alter the entire mid-year shopping season.

The Amsterdam-headquartered startup has only been out of stealth for eight months, but already has 350 employees, production sites on four continents and is reportedly valued at nearly $1.7 billion

There’s a problem that every major enterprise AI implementation encounters at some point: the gap between a convincing demo and a working system in production. Models hallucinate. Break integrations. Compliance requirements vary by country.

Local languages ​​do not behave as US-centric training data suggests. The organizations best positioned to close this gap, the argument goes, are not those with the best models, but those with the most people on the ground.

This thesis is the basis of Wonderful, the enterprise AI agent platform founded in early 2025 by Bar Winkler and Roey Lalazar.

The company has raised $150 million in a Series B round led by Insight Partners with participation from existing backers Index Ventures, IVP, Bessemer Venture Partners and Vine Ventures.

The 💜 of EU technology

The latest rumors from the EU tech scene, a story from our wise founder Boris and questionable AI art. It’s free in your inbox every week. Register now!

The raise brings Wonderful’s total disclosed funding to $286 million, a remarkable figure for a company that only came out of obscurity with a $34 million seed funding round in mid-2025 and then raised $100 million in Series A funding in November of that year.

Wonderful is headquartered in Amsterdam, with Israeli founders and a model based on local operations teams embedded in client organizations. The company says it now operates in more than 30 countries across Europe, the Middle East, Asia Pacific and Latin America, serving companies in telecommunications, financial services, manufacturing and healthcare.

With the new capital, the number of employees will increase from 350 to around 900 by the end of the year.

The company’s core product is an enterprise AI agent platform that is designed to be model agnostic and continuously evaluates and selects AI models for each use case.

Agents handle customer-facing workflows via voice, chat and email, as well as internal workflows such as employee onboarding, compliance and IT support.

What sets Wonderful’s model apart is the deployment layer: Instead of selling software and letting customers do the integration themselves, the company embeds local teams into enterprise environments to manage rollout, integration, and post-deployment optimization.

“In 2026, companies will decide who to work with to implement AI across their organizations, and those decisions will depend on who can provide deep integrations across complex infrastructures and tailor solutions to each organization’s unique environment.” said Bar Winkler, CEO and co-founder of Wonderful.

“We have built our platform and operating model on this reality, and the demand we are seeing globally reflects this.”

The company says more than 70% of companies that start with a single use case expand to additional workflows within three months. Bar Winkler attributes this bond dynamic to Wonderful’s practice of building a common architecture for a company’s core systems from the start. Once this foundation is in place, the activation of new use cases will become faster and faster.

Wonderful also claims measurable operational results from production deployments: up to 60% reduction in processing times, containment rates of over 80%, and millions of dollars in annual efficiency gains for individual customers. These figures are not independently verified.

“Over 70% of companies that start with a single use case expand to additional workflows within the first three months,” added Winkler. “This expansion is possible because we have created a common foundation for all core systems from day one.”

“Wonderful creates trust and deep partnerships within complex organizations at a critical moment for the market,” said Jeff Horing, managing director of Insight Partners. “We believe the team’s combination of platform strength and execution position is wonderful as a strong enterprise partner in today’s ecosystem.”

Lalazar, the company’s CTO, formulated the goal more broadly. “We are deploying agents across all business functions while pioneering the next generation of application layers that will transform the way businesses work,” he said.

The market for enterprise AI agents is crowded and growing. Salesforce’s Agentforce, ServiceNow’s AI platform, and a number of better-funded standalone startups all follow the same budget line.

Wonderful’s differentiation rests on the bet that local delivery teams and multilingual agents will be crucial in markets where U.S.-centric platforms struggle, and that the structural complexity of global companies is effectively their moat.

After eight months in secret, the bet appears to be attracting capital. Whether it will hold up at scale is the question this round of funding revolves around.

We were there. You spend twenty minutes creating the perfect prompt, an AI spits out a beautiful Instagram graphic—and then you realize the font is slightly off and the background color clashes with your brand. What now? Ask and pray again? This is essentially the state of AI-generated design: beautiful results, no editability.

Canva is expressing this with Magic Layers, a new feature now available in beta in the US, UK, Canada and Australia.

Canva

How do Magic Layers work?

Here’s how it works: Insert any flat PNG or JPG format, and the tool reverse-engineers it by pulling out text, objects, backgrounds, and grouped shapes into individual layers that you can actually touch.

Your locked, untouchable JPEG is closer to a real working file; Move things around, swap colors, and fix the slogan. No new beginning.

For a solopreneur trying to turn a design into a multi-channel campaign, this matters. The text remains alive and editable and is not burned in; The layout hierarchy is automatically retained.

Manisha Priyadarshini / Digital Trends

How Canva’s New Feature Helps Everyday Users?

A small business can take an AI-generated advertising image and adapt it for Instagram, a flyer, and an email header—without touching Photoshop or calling a designer.

It is best suited for graphic designs and illustrations. Throw a photorealistic image at it and the results vary, which is honestly fair enough for a beta.

Magic Layers didn’t appear out of nowhere – it fits perfectly into a number of big changes Canva has been making recently. The Canva design model is their own AI trained on design logic, not just producing pretty images.

Canva Sheets took a hard look at spreadsheets and asked why they have to be so miserable – packing data into visual layouts and AI formulas. Video 2.0 rebuilt the editor from scratch.

Three quite different products; A pretty obvious line: AI creates; Canva wants to be where you actually do the work.

Laut IBMs „Cost of a Data Breach Report“ sanken die weltweiten durchschnittlichen Kosten einer Datenschutzverletzung im Jahr 2025 auf 4,44 Millionen US-Dollar, ein Rückgang um 9 Prozent und der erste Rückgang seit fünf Jahren. Oberflächlich betrachtet sieht das nach einem Fortschritt aus. Sicherheits-KI und Automatisierung zahlen sich endlich aus, indem sie die Erkennungszeiten verkürzen und den Ermittlungsaufwand verringern.

Aber die Schlagzeilenzahl verschleiert eine unbequemere Realität. Für Unternehmen mit umfassender Automatisierung sind die Kosten für gemeldete Sicherheitsverletzungen um fast 1,9 Millionen US-Dollar niedriger als für Unternehmen, die auf manuelle Prozesse angewiesen sind. Die Kluft zwischen Spitzenreitern und Nachzüglern schließt sich nicht – sie vergrößert sich. Und genau die KI-Tools, die diese Einsparungen vorantreiben, führen zu einer neuen Risikokategorie, die Aufsichtsbehörden, Versicherer und Vorstände nicht länger ignorieren können.

Das Automatisierungsparadoxon

Sicherheitszentralen haben die KI mit der Dringlichkeit angenommen, mit der eine Branche keine Analysten mehr hat. Die Burnout-bedingten Abwanderungsraten liegen in vielen SOC-Teams bei über 25 Prozent pro Jahr und gehören zu den höchsten in der IT. Der Austausch eines ausgebildeten Analysten dauert in der Regel sechs bis zwölf Monate. Die Rechnung ist brutal: Unternehmen können ihre Resilienz nicht durch Personalbeschaffung erreichen.

Die Automatisierung sollte dieses Problem lösen. Und in engen, klar definierten Arbeitsabläufen, Alarmtriage, Protokollkorrelation und sich wiederholenden Anreicherungsaufgaben – das ist der Fall. Der Nextgen 2025/2026 Cybersecurity Trends Report schätzt, dass die Branchentelemetrie im Jahr 2025 308 Petabyte über mehr als vier Millionen Identitäten, Endpunkte und Cloud-Assets erreichte und fast 30 Millionen Ermittlungshinweise lieferte. Analysten bestätigten nur rund 93.000 echte Bedrohungen von diesem Berg, was einer Trefferquote von lediglich 0,3 Prozent entspricht. Ohne Automatisierung wäre allein das Volumen nicht zu bewältigen.

Das 💜 der EU-Technologie

Die neuesten Gerüchte aus der EU-Tech-Szene, eine Geschichte unseres weisen Gründers Boris und fragwürdige KI-Kunst. Es liegt jede Woche kostenlos in Ihrem Posteingang. Jetzt anmelden!

Doch Gartners Hype-Zyklus 2025 für Sicherheitsoperationen stellt KI-SOC-Agenten auf den Höhepunkt überhöhter Erwartungen und warnt davor, dass die Behauptungen immer noch die nachhaltige, messbare Verbesserung übertreffen. Die anfängliche Einführung führt häufig zu zusätzlichem Aufwand, bevor er ihn verringert. Fehlalarme und Halluzinationen bleiben echte Betriebsrisiken. Und Kostenmodelle schränken häufig den breiten Einsatz über alle SOC-Rollen hinweg ein.

Das Paradox liegt auf der Hand: Unternehmen brauchen KI, um mit der Datenflut fertig zu werden, aber unkontrollierte KI bringt genau die blinden Flecken mit sich, die sie eigentlich beseitigen sollte. Der Bericht von IBM aus dem Jahr 2025 ergab, dass Schatten-KI, bei der Mitarbeiter nicht genehmigte generative KI-Tools zur Verarbeitung sensibler Daten nutzen, durchschnittlich 670.000 US-Dollar an den Kosten für Sicherheitsverletzungen verursachte, sofern vorhanden. Erstaunliche 97 Prozent der angegriffenen Organisationen, bei denen es zu einem KI-bezogenen Sicherheitsvorfall kam, verfügten nicht über angemessene KI-Zugriffskontrollen. Unterdessen gaben 63 Prozent der befragten Organisationen zu, dass sie überhaupt keine KI-Governance-Richtlinien haben.

Die Implikation ist stark. Automatisierung ohne Governance verringert das Risiko nicht, sondern verteilt es neu. Und in einem regulatorischen Umfeld, das zunehmend Transparenz erfordert, ist unkontrollierte KI im SOC nicht nur eine technische Belastung. Es handelt sich um eine Compliance-Belastung.

Wenn Alarmmüdigkeit zum Angriffsvektor wird

Die menschlichen Kosten sind messbar und gehen weit über die Budgetgrenzen hinaus. Im Nextgen-Bericht zitierte Studien zeigen, dass SOC-Teams routinemäßig bis zu 30 Prozent der eingehenden Warnungen ignorieren oder ablehnen – nicht aus Fahrlässigkeit, sondern aus Notwendigkeit. Wenn jede Warnung gleich aussieht und der Kontext fragmentiert über nicht verbundene Konsolen eintrifft, sind erfahrene Analysten gezwungen, ihre Sichtung eher auf ihr Instinkt als auf Beweise zu stützen.

Die Folgen variieren je nach Sektor, aber das Muster wiederholt sich. Im Gesundheitswesen, das mit 7,42 Millionen US-Dollar pro Vorfall und 279 Tagen für die Eindämmung immer noch die kostspieligste Branche für Verstöße ist, ist Alarmmüdigkeit nicht nur ein IT-Problem. Der ENISA-Datensatz zu 215 Vorfällen im Gesundheitswesen zwischen 2021 und 2023 ergab, dass es sich bei 54 Prozent um Ransomware handelte, wobei Patientendaten in 30 Prozent der Fälle das Hauptziel waren. Krankenhäuser haben über umgeleitete Krankenwagen und verzögerte Operationen berichtet, die direkt auf überlastetes Personal und verstopfte Erkennungsleitungen zurückzuführen sind.

Im verarbeitenden Gewerbe und im Energiesektor, wo die Durchsetzung von NIS2 im Jahr 2025 begann, kann ein einziger Ausfalltag in einer Hochdurchsatzanlage Millionen von Euro kosten. Angreifer nehmen zunehmend industrielle Steuerungssysteme ins Visier, indem sie sich durch schlecht segmentierte IT-Netzwerke bewegen und genau die Art mehrdeutiger, kontextabhängiger Warnungen ausnutzen, die überforderte Analysten gerne abtun.

Die Finanzdaten untermauern diesen Punkt. Im Jahr 2025 beliefen sich die Verstöße, die in weniger als 200 Tagen auftraten, auf durchschnittlich 3,87 Millionen US-Dollar, während sich die Verstöße über diesen Schwellenwert hinaus auf durchschnittlich 5,01 Millionen US-Dollar beliefen. Vorfälle in mehreren Umgebungen, die sich gleichzeitig über Cloud-, SaaS- und lokale Infrastrukturen erstreckten, waren mit durchschnittlich 5,05 Millionen US-Dollar und Lebenszyklen von nahezu 276 Tagen sogar noch kostspieliger. Die Betriebsumgebung bestimmt die Komplexität und Komplexität bestimmt die Kosten.

Die Lehre aus dem Jahr 2025 ist, dass das reine Datenvolumen nur zunehmen wird, aber die Teams, die Erfolg haben, sind diejenigen, die Korrelation und Anreicherung als architektonische Notwendigkeiten und nicht als optionale Add-ons betrachten.

Europas regulatorische Konvergenz

Drei Regulierungsrahmen vereinen sich nun auf eine einzige Forderung: Belastbarkeit kontinuierlich nachweisen und nicht erst im Nachhinein melden.

Der Digital Operational Resilience Act (DORA), der im Januar 2025 in der gesamten EU in Kraft trat, definiert die Cybersicherheit für Finanzdienstleistungen neu und richtet sich dabei auf die betriebliche Belastbarkeit bei schwerwiegenden IT-Störungen. Die Berichtspflicht ist das störendste Element – ​​Institutionen müssen innerhalb von Stunden Vorfallberichte einreichen, die durch forensische, prüfungstaugliche Beweise untermauert sind. Protokolle müssen digital signiert und mit einem Zeitstempel versehen sein, um die Prüfung durch die Aufsichtsbehörden Monate später zu überstehen.

Die NIS2-Richtlinie, die 2024–2025 europaweit in nationales Recht umgesetzt wurde, erweiterte den Regulierungsrahmen von sieben Sektoren auf achtzehn wesentliche und wichtige Sektoren. In Rumänien wurde es als Gesetz 124/2025 umgesetzt, das die Fertigung erstmals ausdrücklich als regulierten Sektor benennt und Produktionsstätten dazu zwingt, Compliance-Rahmenwerke einzuführen, die denen von Krankenhäusern und Banken ebenbürtig sind. Gemäß NIS2 sind Vorstände direkt rechenschaftspflichtig, wobei die Strafen unter anderem Geldstrafen und den Ausschluss von der Übernahme von Vorstandsposten in der EU umfassen.

Und dann ist da noch das EU-KI-Gesetz, dessen wesentlichste Verpflichtungen am 2. August 2026 in Kraft treten. Hochriskante KI-Systeme, eine Kategorie, die viele Sicherheitsautomatisierungstools umfasst, müssen die Einhaltung von Anforderungen in Bezug auf Risikomanagement, Datenverwaltung, technische Dokumentation, Transparenz, menschliche Aufsicht, Genauigkeit, Robustheit und Cybersicherheit nachweisen. Anbieter müssen technische Maßnahmen gegen Datenvergiftung, Modellumgehung und gegnerische Angriffe implementieren.

Für globale Finanzkonzerne vervielfacht sich die Komplexität. Ein einzelner Verstoß kann eine gleichzeitige Meldung gemäß DORA, DSGVO und nationalen Rahmenwerken erfordern, jeweils mit unterschiedlichen Formaten und Fristen. Für Hersteller, die neu in den Geltungsbereich von NIS2 fallen, ist die Herausforderung sogar noch grundlegender: Vielen fehlt die Werkzeuginfrastruktur, um überhaupt Nachweise auf Compliance-Niveau zu erstellen, geschweige denn unter Zeitdruck.

Zusammen schaffen diese drei Frameworks ein regulatorisches Umfeld, in dem Cybersicherheits-KI nicht einfach nur effektiv sein kann – sie muss überprüfbar, erklärbar und kontrollierbar sein. Die Frage, vor der Unternehmen stehen, lautet nicht mehr: „Wie sicher sind wir?“ aber „können wir es den Aufsichtsbehörden innerhalb weniger Stunden nachweisen?“. Für Organisationen, die Plattformen evaluieren, die für dieses regulatorische Umfeld entwickelt wurden, bietet ein aktueller Vergleich europäischer SIEM-Anbieter zusätzlichen Kontext.

Das Argument für kontrollierte Autonomie

Diese regulatorische Konvergenz verändert das Aussehen einer guten Sicherheitsarchitektur. Die Branche verlagert sich von der regelbasierten Automatisierung – bei der Playbooks vorgegebene Schritte ausführen – hin zu etwas, das man kontrollierte Autonomie nennen könnte: halbautonome SOC-Operationen mit integrierten Compliance-Schutzmaßnahmen.

In einem Modell der kontrollierten Autonomie ersetzt KI nicht das menschliche Urteilsvermögen. Es engt den Entscheidungsspielraum ein. Die Korrelation erfolgt bei der Aufnahme, wodurch Dutzende fragmentierter Warnungen in einem einzigen angereicherten Fall mit vollständigen Prüfnachweisen zusammengefasst werden.

Das UEBA-Scoring ordnet anomale Identitäten und Vermögenswerte nach Risiko, sodass sich Analysten auf das Wesentliche konzentrieren können, anstatt sich durch das Chaos zu kämpfen. Und jeder Untersuchungszeitplan dient gleichzeitig als Compliance-Artefakt, digital signiert, im Rahmen abgebildet und bereit für den Export durch die Regulierungsbehörden.

Das Architekturprinzip ist schlank: Jeder Sicherheitsfall ist gleichzeitig ein Compliance-Fall. Analysten führen einmal eine Untersuchung durch, und das System erstellt sowohl operative Ergebnisse als auch für die Regulierungsbehörde bereitstehende Berichte. Dadurch wird die Duplizierung vermieden, die Unternehmen mit separaten SIEM-, SOAR- und Compliance-Tools belastet, die jeweils Kosten, Latenz und Integrationsaufwand verursachen.

Europäische Plattformen basieren zunehmend auf dieser Philosophie. Das in Rumänien ansässige Unternehmen Nextgen Software hat beispielsweise seine CYBERQUEST-Plattform entwickelt, um Erkennung, Untersuchung und Compliance-Berichte in einem einzigen Workflow zu vereinen, sodass jeder angereicherte Fall automatisch den Prüfpfad DORA und NIS2-Anforderung generiert. Sein agentenloses OT-Überwachungsmodul schließt eine Lücke, die für Hersteller und Versorgungsunternehmen von Bedeutung ist: Einblick in industrielle Steuerungssysteme ohne den Einsatz aufdringlicher Endpunktagenten. Ähnliche Konvergenzbemühungen sind in der gesamten europäischen Anbieterlandschaft sichtbar, von nordischen SIEM-Anbietern, die Compliance-fähige Exporte erstellen, bis hin zu von Deutschland geführten Initiativen, die ISO 27001- und NIS2-Zuordnungen direkt in die Erkennungslogik einbetten.

Vom Assistenten zum Agenten – sorgfältig

Die nächste Grenze ist der Übergang von KI-Assistenten zu KI-Agentensystemen, die nicht nur die nächsten Schritte vorschlagen, sondern aktiv Erkennungs-, Untersuchungs- und Reaktionsabläufe ausführen. Es ist ein Übergang, dem die Branche mit einer Mischung aus Ehrgeiz und Vorsicht entgegengeht.

Vlad Gladin, CTO von Nextgen Software, beschreibt diese Entwicklung in praktischen Worten: „Unsere Cyber Minds AI Personas entwickeln sich von beratenden Assistenten zu kontextbewussten Ermittlungsagenten. Anstatt nur eine Antwort zu empfehlen, können diese Agenten Telemetriedaten über Identitäts-, Netzwerk- und Endpunktdaten in Echtzeit korrelieren, vorläufige forensische Analysen durchführen und Analysten eine erweiterte Untersuchungserzählung präsentieren, nicht eine Warteschlange getrennter Warnungen. Das Ziel besteht nicht darin, den Analysten aus der Situation zu entfernen.“ Schleife, aber um sicherzustellen, dass der Kontext bereits zusammengesetzt ist, wenn sie eingreifen.“

Dies spiegelt die allgemeine Entwicklung der Branche wider. Gartner empfiehlt, KI-SOC-Agenten als Tools zur Workflow-Erweiterung und nicht als autonomen Ersatz zu behandeln, wobei der Schwerpunkt auf der Aufrechterhaltung der menschlichen Aufsicht liegen sollte. Die Sorge ist berechtigt: Überautomatisierung birgt Risiken, wenn Agenten auf der Grundlage fehlerhafter Annahmen handeln, und die meisten aktuellen Anwendungsfälle bleiben eng und aufgabenspezifisch und nicht durchgängig.

Der kontrollierte Ansatz bedeutet, schrittweise Vertrauen aufzubauen. Beginnen Sie mit der automatisierten Anreicherung und Gehäusemontage. Ebene der UEBA-gesteuerten Priorisierung. Erst dann auf halbautonome Reaktionsmaßnahmen ausweiten – und immer mit Prüfprotokollen, die eine Aufsichtsbehörde oder ein Versicherer im Nachhinein überprüfen kann.

Es gibt einen Grund, warum dieses inkrementelle Modell in Europa besonders großen Anklang findet. Die Regulierungslandschaft des Kontinents belohnt die nachweisbare Kontrolle über die Rohkapazitäten. Ein KI-Agent, der tausend Warnungen pro Stunde selektieren kann, ist beeindruckend; Ein KI-Agent, der tausend Alarme pro Stunde selektieren und für jeden einen DORA-konformen Vorfallzeitplan erstellen kann, ist bankfähig. Die kommerzielle Logik und die Regulierungslogik konvergieren bei denselben architektonischen Anforderungen.

Was 2026 verlangt

Die Organisationen, die für 2026 am besten aufgestellt sind, sind nicht unbedingt diejenigen mit der fortschrittlichsten KI, aber diejenigen, die nachweisen können, dass ihre KI vertrauenswürdig ist. In einem Umfeld, in dem DORA forensische Beweise innerhalb weniger Stunden verlangt, NIS2 Vorstände persönlich haftbar macht und das EU-KI-Gesetz eine nachweisbare Steuerung von Hochrisikosystemen erfordert, ist das eigentliche Unterscheidungsmerkmal nicht die Geschwindigkeit der Erkennung, sondern die Geschwindigkeit des nachweisbaren Vertrauens.

Dies bedeutet, dass Compliance keine ergänzende Aufgabe bleiben darf, die vierteljährlich von einem separaten Team durchgeführt wird. Es muss in den Arbeitsablauf von der Erkennung bis zur Lösung eingebettet sein und automatisch als Nebenprodukt der Vorfallbearbeitung generiert werden. Plattformen, die revisionssichere Beweise als natürliches Betriebsergebnis liefern, anstatt von Analysten eine nachträgliche Rekonstruktion zu verlangen, werden den neuen Standard setzen.

Die Cybersicherheitsbranche hat im letzten Jahrzehnt einen Wettlauf um die Automatisierung verbracht. Im Jahr 2026 verlagert sich das Rennen auf die Steuerung dieser Automatisierung und beweist gegenüber Regulierungsbehörden, Versicherern und Vorständen, dass die Maschinen, die das Netzwerk verteidigen, selbst verantwortlich sind. Die Gewinner werden nicht die Organisationen mit der meisten KI sein. Sie werden diejenigen sein, deren KI zeigen kann, dass sie funktioniert.

Researchers have developed a unique smart pillow designed to stream music and podcasts, giving users a new way to relax without relying on smartphones. The experimental device, developed by a team at the University of Glasgow, aims to promote healthier nighttime habits by reducing the temptation to endlessly scroll through social media before bed.

The project addresses growing concerns about “doomscrolling” – the habit of constantly browsing negative or overwhelming content on smartphones, especially late at night. By providing a gentler and more calming way to consume audio content, researchers hope to help people unwind more effectively and improve their overall digital well-being.

A pillow to relax

The smart pillow integrates built-in audio technology, allowing users to stream music, podcasts or other audio content directly through the pillow itself. Instead of using headphones or a phone screen, users can simply place their head on the pillow or hug to hear the sound.

Unsplash

The device transmits audio vibrations through the structure of the cushion. This method allows the sound to be clearly heard by the user while minimizing disruption to others nearby. Because the sound is transmitted through physical contact with the pillow, it creates a more personalized listening experience compared to traditional speakers.

Researchers say the design encourages people to focus on soothing sound rather than constantly interacting with their phones.

Doomscrolling is a modern menace

“When people doom scroll for hours, they are also sedentary for a long time. The effects are enormous and problematic,” notes Dr. Aditi Nerurkar, lecturer at Harvard Medical School. The analysis, published in Harvard Health and citing numerous research papers, links doomscrolling to increased existential anxiety and reduced performance, alongside the obvious physical tool of sedentary behavior.

A 2025 study by the American Heart Association found that most Americans die from heart disease each year, and clearly suggested that a sedentary lifestyle “is associated with an increased risk of fatal cardiovascular disease.” Unsurprisingly, doomscrolling on phones makes up for a large portion of those sedentary minutes or hours.

Another study published in the journal Applied Research in Quality of Life linked doom scrolling to compulsive addictive behavior, psychological distress and reduced mental health. Separately, a 2024 research report in the journal Computers in Human Behavior Reports found direct links between doomscrolling and higher levels of stress. Music, on the other hand, is known to have a calming effect on the individual.

Digital wellbeing is important

Digital wellbeing has become an increasingly important issue as smartphones dominate everyday life. Many tech companies are introducing tools to limit screen time or reduce distractions, but physical solutions like this pillow offer a different approach.

Sleep Unsplash

By combining convenience and technology, the device could offer people a more natural way to disconnect from their screens without completely sacrificing entertainment or information. Researchers also believe the pillow could be beneficial for people who use audio content as part of their bedtime routine, including listening to podcasts, meditation guides or relaxing music.

The smart pillow is currently a research prototype

This means that it is not yet available as a commercial product. However, the project shows how everyday objects could evolve into tools that support healthier technology habits.

Future versions could include additional features such as wireless connectivity, customizable audio settings or integration of sleep-focused content. If the concept proves successful, it could also inspire similar designs that incorporate technology into everyday household items.

For now, the smart pillow marks an interesting direction for technology design – one that prioritizes comfort, well-being and reduced screen dependency rather than adding yet another device to people’s already crowded digital lives.

A Puglia-based startup founded by the man behind aircraft manufacturer Blackshape has closed a pre-seed round to build software-defined ships and maritime AI. The ocean is the last large physical environment that is not yet controlled by software.

Mirai Robotics, a startup headquartered in Puglia in southern Italy, wants to change that. The company has closed a €3.9 million (approximately $4.2 million) pre-seed round to develop autonomous surface vessels and a maritime reconnaissance platform designed for continuous uncrewed operations in coastal and offshore environments.

The round was led by Primo Ventures, Techshop and 40Jemz Ventures, with participation from Italian and international angel investors.

Primo Ventures, whose chairman and general partner Gianluca Dettori commented on the deal, is an Italian early-stage VC firm that manages around €438 million in digital, space, healthcare and climate technology funds. It is one of the most active seed investors in Italy.

What Mirai is building

At the heart of the Mirai platform are software-defined autonomous surface vehicles designed for what the company calls dock-to-dock autonomy: the ability to complete entire missions without human intervention from departure to return. The ships combine advanced sensor systems, autonomous navigation, remote monitoring tools and integrated security layers.

The company has already developed two autonomous vehicles for intelligence, surveillance, reconnaissance and maritime patrol missions, operating either as standalone units or within coordinated fleets.

In addition to its own hardware, Mirai also develops modular autonomy and control systems that can be integrated into third-party vessels, meaning shipyards, industrial operators and public bodies can adopt autonomous technology without decommissioning existing fleets.

Underlying both are a proprietary maritime reconnaissance and mission management platform that provides what Mirai describes as persistent domain awareness: the ability to monitor maritime environments, coordinate robotic assets and maintain operational control over complex conditions over extended periods of time.

“The ocean is one of the last major physical infrastructures not yet controlled by software,” Mirai CEO Luciano Belviso said in a statement. “Autonomy is the key to finally making the oceans safe and usable, unlocking enormous resources and addressing critical security challenges. But it must be implemented through systems capable of operating continuously and safely in extreme environments.” This is a technological and industrial challenge that requires a true robotic laboratory approach.”

The founders

The three co-founders bring unusually different qualifications for a young robotics startup. Belviso founded Blackshape Aircraft in Puglia in 2009, a carbon fiber aircraft manufacturer that builds high-performance two-seaters for both the recreational and military training markets and is now part of the Angel industrial holding group.

He holds degrees in aerospace engineering, mechanical engineering and space law from institutions such as the Polytechnic of Turin, the EPFL in Lausanne and the Université Paris XI.

Luca Mascaro is the founder and president of Sketchin, a Swiss-based strategic design studio that became part of BIP Group, the Italian management consultancy, following a majority acquisition in 2016. Mascaro remained at Sketchin as founder and president after the deal. At Mirai he brings experience in building technology-oriented service companies at a European level.

Davide Dattoli is the founder and CEO of Talent Garden, one of Europe’s largest education and technology community networks, operating in 12 markets and training around 25,000 professionals annually. He was named to the Forbes “30 Under 30 Europe” list and is a venture founder at the Italian Founders Fund.

The market context

The strategic logic for maritime autonomy is not difficult to find. According to the European Commission, Europe’s blue economy – the various economic sectors that depend on or interact with the sea, including shipping, fishing, offshore energy and port operations – is worth over €750 billion a year.

In addition, the company faces increasing pressures: rising operating costs, a deepening workforce crisis as experienced maritime professionals retire, and an increasing need for persistent monitoring of infrastructure, including submarine cables, offshore wind farms and energy platforms.

The dual-use approach is also important. Autonomous ships for patrol and ISR missions stand at the intersection of civilian maritime operations and defense technology, a sector that is attracting a growing share of European venture capital as governments across the continent increase their defense budgets and seek sovereign capabilities in monitoring critical infrastructure.

Mirai is based in Puglia, which the founders describe as an ideal location for their ambitions as it lies at the intersection of Mediterranean maritime activity, industrial manufacturing tradition and academic research institutions.

The company says the funding will accelerate its technology stack, expand its engineering team and support pilot deployments with industrial and institutional partners.

“The maritime space is at an inflection point,” Primo Ventures’ Dettori said in a statement. “We are dealing with a huge economy that is still based on operating models developed decades ago.

The lack of human capital, thousands of unfilled positions, an aging workforce and increasing operational risks alone make the status quo untenable. What Mirai Robotics is developing is not just automation; It is the foundational infrastructure layer that enables the blue economy to scale safely and efficiently.”

Researchers have developed a new system that uses smartphones instead of specialized motion tracking hardware to create more realistic avatar movements in virtual reality (VR). The technology aims to reduce the need for expensive VR equipment while making immersive digital interactions more accessible to a wider range of users.

The system was developed by researchers who wanted to solve a long-standing challenge in virtual environments: capturing natural body movements and accurately translating them into digital avatars. Traditional VR setups often rely on dedicated motion capture suits, cameras or tracking sensors distributed around the room. Although these systems are effective, they can be expensive and complicated to install.

The new approach uses something that many people already own – a smartphone. Using the motion sensors built into phones, researchers developed a way to track body movements and reproduce them in a virtual environment with surprising accuracy.

At the heart of the system is a method that interprets data from a phone’s accelerometer, gyroscope and orientation sensors

These sensors measure how the device moves and rotates in space. The system then feeds this information into software that predicts the movement of the user’s body and translates it into a digital avatar.

VR headset Unsplash

In practice, a user can place a smartphone on their body – for example, in a pocket or strapped to their torso – while wearing a standard VR headset. The phone collects motion data when the user walks, turns, or gestures. Algorithms process this data to animate the avatar in real time, allowing it to mimic natural body movements without the need for a full motion capture setup.

Researchers say the technology can produce more realistic avatar behavior, especially compared to simple VR systems that only track the head and hands. When avatars move more naturally, immersion in virtual environments can be greatly improved.

The progress is important because realistic avatars are essential for many new applications

Virtual meetings, online social platforms, collaborative design tools, and remote training environments all rely on avatars that accurately represent human movements.

Without full-body motion tracking, avatars often appear stiff or unnatural. This can reduce the sense of presence that makes virtual environments seem compelling. By using smartphones to capture additional body movements, the new system could make digital interactions feel much more natural.

VR headset Unsplash

Another advantage is accessibility. High-end motion tracking systems used in professional VR setups can cost thousands of dollars. A smartphone-based solution dramatically lowers this hurdle, potentially enabling realistic avatars for people using consumer VR headsets.

The technology could also benefit industries outside of gaming and social VR. For example, training simulations in healthcare, education or occupational safety could become more immersive if participants’ movements are more accurately recorded.

Researchers note that the system still has limitations

Since it is based on a small number of sensors compared to complete motion capture devices, it cannot yet capture every subtle movement of the body. However, the algorithms can estimate specific movements by analyzing patterns in the data.

Future developments will focus on improving these predictions and integrating the system with additional wearable sensors if necessary. Researchers are also exploring ways to make the system work seamlessly with existing VR platforms and headsets.

As the technology continues to improve, it could play a key role in bringing realistic avatars to millions of users without the need for expensive equipment. As virtual worlds become more prevalent – ​​from online collaboration spaces to entertainment platforms – solutions that make immersive technology easier and more affordable could help accelerate adoption across industries and everyday life.