Public concern over big tech companies is growing in Switzerland, according to a new survey by gfs.bern conducted on behalf of the Mercator Foundation Switzerland. A large majority of respondents view major technology firms as primarily profit-driven, while also expressing unease about their broader influence on society and politics.
Survey findings show that 90% of respondents believe big tech companies are mainly motivated by profit, while 94% support stronger protections for children and young people on social media platforms. Concerns extend beyond commercial behaviour, with 84% worried about political influence from the countries where these companies are based and 82% fearing increasing dependence on firms from the United States and China.
Overall perceptions in Switzerland remain mixed: 21% of respondents express a positive view of big tech companies, 40% hold a neutral stance, and 38% report negative impressions. Similar attitudes have been observed across Europe, where surveys in countries such as France and Germany indicate that many citizens consider existing regulatory frameworks insufficient.
Despite concerns about corporate influence, attitudes towards digitalisation itself remain broadly positive. Around 58% of respondents see digitalisation as beneficial overall, and 53% believe it offers personal advantages. However, only 48% think it benefits society as a whole, while 46% perceive its impact on democratic processes as negative.
A strong majority expects public institutions to take on greater responsibility for managing digital transformation. Around 88% support government efforts to ensure transparency in AI decision-making, while 86% want human oversight in critical situations. High levels of trust in Swiss authorities suggest public backing for a more active state role in shaping digital policy and safeguarding democratic values.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
In 2019, Facebook CEO Mark Zuckerberg announced Facebook Horizon, a VR social experience that allows users to interact, create custom avatars, and design virtual spaces. Zuckerberg saw the platform, later renamed Horizon Worlds, as the beginning of a new era of VR social networks, with users trading face-to-face interactions for digital ones.
To show his confidence in VR, Zuckerberg rebranded Facebook Inc. as Meta Platforms Inc. in October 2021, illustrating the company’s shift toward the metaverse as a broad virtual environment intended to integrate social interaction, work, commerce, and entertainment. Building on this new vision, Meta’s ambitions expanded beyond social interaction and entertainment, with the development roadmap including virtual real estate purchases and collaboration in virtual co-working spaces.
Fast forward to 17 March 2026, and the scale of Meta’s retreat from the metaverse vision has become unmistakable. In an official update, the company said it was ‘separating’ VR from Horizon so that each platform could grow with greater focus, while also making Horizon Worlds a mobile-only experience. Under the plan, Horizon Worlds and Events would disappear from the Quest Store by 31 March 2026, several flagship worlds would no longer be available in VR, and the Horizon Worlds app itself would be removed from Quest on 15 June 2026, ending VR access to Worlds altogether.
Yet Meta soon reversed part of the decision. In an Instagram Stories Q&A, CTO Andrew Bosworth said Horizon Worlds would remain available in VR after user backlash. Even so, the greater shift remained unchanged: Horizon Worlds was no longer a flagship VR project, but a much narrower product that reflected a clear contraction of Meta’s original metaverse ambition.
As it stands, Meta’s USD 80 billion investment seems less like a gateway to a new socio-technological era and more like one of the most expensive strategic miscalculations of the 21st century. The sunsetting of Horizon Worlds was certainly not a decision made on a whim, which begs the question: Why did the metaverse fail in the first place? Does it have a future in the AI landscape, and what does its retreat say about the politics of designing the future through corporate platforms?
Metaverse’s mainstream collapse
The most obvious reason for the metaverse’s failure was that it never became a mainstream social space. Meta’s strategy rested on the belief that large numbers of people would start using immersive virtual worlds as a normal setting for interaction, entertainment, and creative activity. The shift never happened at the scale needed to sustain the company’s ambitions.
One reason was friction. VR headsets were less practical than phones, more isolating than social media, and harder to integrate into everyday routines than the platforms people already used to communicate. Entering the virtual world required extra time, extra hardware, and openness to adapt to a different social environment. Most digital habits, however, are built around speed, familiarity, and ease of access.
Meta’s own March 2026 decision makes that failure difficult to deny. A company still convinced that immersive social VR was on its way to becoming mainstream would not have moved Horizon Worlds away from Quest and towards mobile. The shift suggested that the metaverse had failed to move from technological promise to everyday social practice.
Metaverse’s failure was not just one of convenience. It also struggled because it was never presented simply as a new digital space. It was framed as a future built largely on Meta’s own terms, with access tied to the company’s hardware, platforms, rules, and wider ecosystem. Such decisions made the metaverse feel less like an open evolution of the internet and more like a tightly managed corporate environment.
The distinction mattered because Meta was not merely launching another product. It was promoting a vision of how people might one day work, socialise, shop, and create online. Yet the more expansive that vision became, the more obvious it was that the system behind it remained closed and centralised. A future digital environment is harder to embrace when a single company controls the devices, spaces, distribution, and boundaries of participation.
Meta’s handling of Horizon Worlds clearly exposed that tension. The company could remove features, reshape access, alter incentives, and redirect the platform from the top down. Such a level of control may be standard for a private platform, but it sits uneasily with claims about building the next phase of digital life. In that sense, the metaverse failed not only because people were unconvinced by VR, but because its version of the future felt too corporate, too enclosed, and too disconnected from the openness people still associate with the internet.
Metaverse’s economic contradiction
The metaverse did not fail only as a social project. It also became increasingly difficult to justify on economic grounds. Meta spent heavily on Reality Labs while generating only limited returns from those investments. In its 2025 annual filing, the company said Reality Labs had reduced overall operating profit by around USD 19.19 billion for the year, while warning that similar losses would continue into 2026.
Losses on that scale might still have been acceptable if the metaverse had shown clear signs of momentum. However, there was little evidence of mass adoption, strong retention, or a durable path to monetisation. Virtual land, digital goods, branded experiences, and immersive workspaces never developed into the economic base of a new internet layer.
Instead, the metaverse began to look less like a future growth engine and more like a costly experiment with uncertain returns. The gap between spending and payoff became harder to ignore, especially as Meta continued to frame the metaverse as a long-term strategic priority. What used to be sold as the company’s next major frontier was increasingly difficult to justify in commercial terms.
The broader strategic context also changed. Meta’s own forward-looking statements pointed to increased hiring and spending in 2026, especially in AI. In practice, this meant the company was no longer choosing between the metaverse and inactivity, but between two competing visions of the future. AI was already delivering tangible gains in product development, infrastructure, and investor confidence.
In that competition for attention and capital, the metaverse lost. Meta’s pullback was also not an isolated case. Microsoft moved away from metaverse-first ambitions as well, retiring the Immersive space (3D) view in Teams meetings, Microsoft Mesh on the web, and Mesh apps for PC and Quest in December 2025. The services were replaced by immersive events in Teams, a narrower offering built around specific workplace functions rather than a broad metaverse vision.
The wider retreat matters because it suggests the problem was not limited to Meta’s execution. Another major tech company also stepped back from standalone immersive environments and turned to more limited, use-specific tools instead. A larger pattern appeared from that shift: grand metaverse narratives gave way to practical features, embedded tools, and industry-specific uses. In that sense, the metaverse has not entirely disappeared, but it did lose its status as the next internet.
Metaverse’s afterlife in the age of AI
The metaverse’s decline does not necessarily imply a complete disappearance. What seems more likely is that parts of it will survive in altered form, detached from the sweeping vision that once surrounded it. Rather than continuing as a standalone digital world meant to transform social life, the metaverse may persist as a set of tools, features, and immersive functions folded into other technologies.
AI is likely to play a role in that transition. It can lower the cost of building virtual environments, speed up avatar creation, automate elements of interaction design, and make digital spaces more responsive. In this sense, AI may succeed where the original metaverse struggled, not by reviving the same vision, but by making parts of it more practical and easier to use.
Such a distinction is important because it shifts the focus from ideology to utility. The metaverse was once marketed as the next stage of the internet, yet its more durable applications now appear to lie in narrower settings where immersion serves a clear purpose. Training, design, simulation, and industrial planning are all contexts in which virtual environments can offer measurable value without becoming a universal social destination.
What might survive, then, is not the metaverse as it was originally imagined, but a smaller set of immersive capabilities embedded in gaming, education, industry, and workplace systems. Avatars, digital agents, simulations, and adaptive virtual spaces may all remain relevant, but as components rather than the foundation of a new social order.
The shift also helps explain the political lesson of the metaverse’s collapse. Large-scale investment, aggressive branding, and executive certainty were not enough to secure public legitimacy. Meta tried to present the metaverse as an inevitable horizon, yet users did not embrace it, markets did not reward it in proportion to the spending, and the company itself eventually narrowed the project it had once elevated into a corporate identity.
In that sense, the metaverse matters even in failure. Its retreat does not simply mark the end of an overhyped product cycle. It also reveals the limits of top-down corporate future-making, especially when private platforms try to define the direction of collective digital life before society has decided whether such a future is either desirable or necessary.
Conclusion
The metaverse failed because it asked too much of users, promised too much to investors, and concentrated too much power in a platform model that never convincingly earned public trust. Meta’s retreat from Horizon Worlds makes that failure difficult to ignore, while Microsoft’s parallel narrowing of immersive ambitions suggests the problem extended beyond one company’s misjudgement.
Immersive VR technologies are unlikely to vanish, and AI may even extend some of their useful applications. Yet the metaverse as a universal social future has largely collapsed under the combined weight of weak adoption, unsustainable economics, and an overly corporate vision of digital life. What remains is not the next internet, but a reminder that the future cannot simply be declared into existence by the companies most eager to own it.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Meta has unveiled its first prescription-optimised AI glasses, expanding its wearable line with Ray-Ban Meta Gen 2 models for everyday vision correction. The launch targets users who already rely on prescription eyewear, offering a more integrated and comfortable experience.
The range includes Blayzer Optics and Scriber Optics with adjustable hinges, nose pads, and temple tips for a better fit. Pre-orders begin at $499 in the United States via Meta and Ray-Ban platforms, with wider availability in optical retailers and select global markets from 14 April.
Alongside the hardware launch, Meta is introducing new frame and lens colour combinations across its Ray-Ban Meta and Oakley Meta collections.
Additional AI-driven features are also rolling out, including hands-free nutrition tracking, WhatsApp message summaries, and improved on-device recall capabilities designed to enhance everyday communication.
Further software updates extend functionality with discreet handwriting input, in-lens navigation across US cities, and expanded media recording tools. The company positions its AI glasses as a multifunctional platform combining vision correction, connectivity, and real-time assistance.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
The US Federal Trade Commission has taken action against OkCupid and Match Group Americas over allegations that the dating app shared users’ personal information, including photos and location data, with an unrelated third party despite privacy promises saying such sharing would not occur without notice or an opportunity to opt out.
According to the FTC’s complaint, OkCupid gave the third party access to personal data from millions of users even though the recipient was not a service provider, business partner, or affiliate within the company’s corporate family. The agency says consumers were not informed and were not given a chance to opt out.
The complaint says the third party sought large OkCupid datasets because OkCupid’s founders were financial investors in that company, despite there being no business relationship with the app. The FTC alleges that OkCupid provided access to nearly 3 million user photos, along with location and other information, without formal or contractual limits on how the data could be used.
Christopher Mufarrige, Director of the FTC’s Bureau of Consumer Protection, said: ‘The FTC enforces the privacy promises that companies make. We will investigate, and where appropriate, take action against companies that promise to safeguard your data but fail to follow through—even if that means we have to enforce our Civil Investigative Demands in court.’
The FTC also alleges that, since September 2014, Match and OkCupid have taken extensive steps to conceal and deny that the apps shared users’ personal information with the data recipient, including conduct the agency says obstructed its investigation. One example cited in the complaint is that, after a news report revealed the third party had obtained large OkCupid datasets, the company told the media and users that it was not involved with that third party.
Under the proposed settlement, OkCupid and Match would be permanently prohibited from misrepresenting how they collect, maintain, use, disclose, delete, or protect personal information, including photos, demographic data, and geolocation data. Restrictions would also cover how they describe the purposes of data collection and disclosure, as well as how they present privacy controls and consumer choices under state privacy laws.
The Commission vote authorising staff to file the complaint and stipulating the final order was 2-0. The FTC filed both in the US District Court for the Northern District of Texas, Dallas Division. The agency notes that a complaint reflects its view that it has ‘reason to believe’ the law has been or is about to be violated, while stipulated final orders carry the force of law only if approved and signed by the district court judge.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Cloudflare has announced two changes to its client-side security offering, making Client-Side Security Advanced available to self-serve customers and offering domain-based threat intelligence at no extra cost to all users on the free Client-Side Security bundle. The update is focused on browser-based attacks that can steal data via malicious scripts without visibly disrupting a website’s normal operation.
Cloudflare says its client-side security system assesses 3.5 billion scripts per day and monitors an average of 2,200 scripts per enterprise zone. According to the company, the product relies on browser reporting, including Content Security Policy signals, rather than scanners or application instrumentation, and requires only that traffic be proxied through Cloudflare.
A central part of the announcement is a new detection pipeline combining a Graph Neural Network (GNN) with a Large Language Model (LLM). Cloudflare says the GNN analyses the Abstract Syntax Tree of JavaScript code to identify malicious intent even when scripts are minified or obfuscated. Scripts flagged as suspicious are then passed to an open-source LLM running on Workers AI for a second-stage semantic assessment intended to reduce false positives.
Cloudflare says the GNN is tuned for high recall to identify novel and zero-day threats, but that false alarms remain a challenge at internet scale. Internal evaluation results cited by the company show that the secondary LLM layer reduced false positives in the JS Integrity threat category by nearly three times across the total analysed traffic, lowering the rate from about 0.3% to about 0.1%. On unique scripts, Cloudflare says the false-positive rate fell from about 1.39% to 0.007%.
The company also describes a recent case involving a heavily obfuscated malicious script named core.js. According to Cloudflare, the payload targeted Xiaomi OpenWrt-based home routers, altered DNS settings, and attempted to change admin passwords. Cloudflare says the script was injected through compromised browser extensions rather than by directly compromising a website, and adds that its GNN detected the malicious structure while the LLM confirmed the intent.
Cloudflare argues that the two-stage design provides structural detection via the GNN and broader semantic filtering via the LLM, enabling the company to lower the GNN decision threshold without sharply increasing alert volume. Every script flagged by the GNN is also logged to Cloudflare R2 for later auditing, which the company says helps it review cases where the LLM overrode the initial verdict.
Domain-based threat intelligence is now being made available to all Client-Side Security customers, including those not using the Advanced tier. Cloudflare says the move is partly a response to attacks seen in 2025 against smaller online shops, especially on Magento, where client-side compromises continued for days or weeks after public disclosure. By extending domain-based signals more broadly, the company says site owners can more quickly identify malicious JavaScript or suspicious connections and investigate possible compromises.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
A new study by CGI.br and NIC.br examines how digital services in Brazil implement age assurance measures. Presented in Brasília during an event on the Digital Child and Adolescent Statute (ECA Digital), the study reviewed 25 popular online services used by children and adolescents.
The study found that most of the services analysed do not apply age checks at the point of registration, including some platforms aimed at adults. According to the release, age assurance usually appears later, when users try to access specific features such as livestreaming or monetisation.
Titled ‘Age assurance practices in 25 digital services used by children in Brazil’, the study analysed governance documents published before the ECA Digital entered into force. From 18 March, the law requires information-society services aimed at children and adolescents in Brazil, or likely to be accessed by them, to adopt effective age-assurance measures and parental supervision.
The study found that 11 of the 25 platforms relied on third-party age-assurance services, particularly social media and generative AI platforms. Official identity document submission was the most common verification method, while selfie-based checks were the most common age-estimation tool. Differences were also found between the minimum ages stated by services and those listed in app stores, and some adult-oriented platforms could still be accessed by younger users with parental consent.
Parental supervision tools were available in 15 of the 25 services, but activation was usually optional and depended on parents or guardians. Transparency also emerged as a weakness: only six services published Brazil-specific reports, and only one explained how its minimum-age policy was applied. Policies were often spread across multiple pages, averaging 22 pages per service, and around 40% of the services provided related information in other languages.
Fábio Senne, General Research Coordinator at Cetic.br | NIC.br, said: ‘One of the study’s central aims was to verify the integrity of the information made available by digital services in Brazil. It is essential that data on age protection be communicated clearly and accessibly, allowing more informed and effective parental supervision.’
Juliana Cunha, manager of the Digital Public Policy Advisory Office at CGI.br | NIC.br, said: ‘This survey was developed to support the debate on implementation of the ECA Digital and to offer a clear understanding of the current landscape. This initiative forms part of a broader set of actions by CGI.br and NIC.br aimed at providing technical evidence to support effective enforcement of the law. Our commitment is to foster a safer and more responsible digital ecosystem for children and adolescents in Brazil.’
The release says the study used as a methodological reference the OECD technical paper ‘Age assurance practices of 50 online services used by children’, published in 2025. Information was collected between 10 and 30 January 2026 from public documents made available by the services in Brazil, totalling 550 pages analysed. The event also marked the launch of TIC Kids Online Brazil 2025, a publication on internet use by children and adolescents aged 9 to 17 in Brazil.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
The International Association of Privacy Professionals (IAPP) has updated its US State Breach Notification Chart, a resource that summarises state breach notification laws across the United States. In an analysis published on 26 March, the IAPP says the revised chart highlights both nationwide coverage and continuing variation in how states define personal information, apply harm thresholds, and trigger reporting duties.
According to the IAPP, all 50 states, the District of Columbia, Guam, Puerto Rico, and the US Virgin Islands now have breach notification laws. California enacted the first state law in 2002, which took effect in 2003, while Alabama was the last state to adopt such a law in 2018. The IAPP says the result is a de facto nationwide framework, but one marked by significant differences across jurisdictions.
A central point in the analysis is that breach notification laws generally use a narrower definition of personal information than more recent comprehensive privacy laws. The IAPP says the original purpose of breach notification was to alert people to the risks of identity theft and financial fraud after a data breach, so laws tend to focus on identifiers such as names combined with Social Security numbers, driver’s licence details, or financial account credentials.
The article contrasts narrower statutes with broader ones. Hawaii’s law is described as among the narrowest, while Illinois and California are presented as having broader definitions that can extend to medical information, health insurance details, biometric data, genetic data, and, in California’s case, some automated licence plate recognition data.
Even so, the IAPP says many state breach laws still do not cover large categories of digital information, such as browsing history, cookie data, IP addresses, cell phone numbers, purchasing records, or complete financial transaction histories where account credentials were not compromised.
Exemptions and scope also vary. The IAPP says most breach notification laws apply broadly to businesses and often to nonprofit organisations, while privacy laws tend to contain more exclusions. The article notes that some states cover state and local government entities directly, while California has a separate breach notification law for governmental bodies. The IAPP also says its chart is focused on laws applicable to the private sector.
Encryption safe harbours appear across the state laws, according to the analysis, with some states also recognising redaction or other protections that render data unreadable or unusable. Attorney general notification requirements also differ. The IAPP says 34 state laws require notice to the state attorney general once certain thresholds are met, with thresholds ranging from 250 affected residents in North Dakota and Oregon to 1,000 in many other states, while some states, such as Connecticut and New York, require notice regardless of the number affected.
Harm thresholds are another area of divergence. The IAPP says about 30 state laws include a harm standard, meaning notice may not be required unless the breach caused, or is likely to cause, harm to affected individuals.
The article describes substantial differences in wording across states, with some referring to ‘reasonable likelihood’ of harm, others to ‘material risk,’ ‘substantial economic loss,’ or misuse of the data, while some states, including California, Georgia, Illinois, Massachusetts, Minnesota, North Dakota, and Texas, require no harm showing at all.
The practical effect, the IAPP argues, is that organisations holding data on residents of multiple states face a complex compliance problem. A data element that triggers notice in one state may not do so in another, and the article says reconciling the different harm standards is effectively impossible. The analysis notes that some organisations may decide to notify if there is doubt, while others may choose to notify only where clearly required.
The IAPP concludes that the absence of a preemptive federal breach notification law leaves entities to navigate overlapping but inconsistent state rules. Its updated chart is presented as a tool to help practitioners track those differences and build awareness of how US state breach notification laws continue to evolve.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
An opinion article published by the International Association of Privacy Professionals says India’s data protection and AI governance environment is facing growing pressure as compliance work around the Digital Personal Data Protection Act (DPDPA) unfolds, court challenges continue, and regulators widen oversight into new sectors. The piece, published on 26 March, is labelled as an opinion article and includes an editor’s note stating that the IAPP is policy neutral and publishes contributed opinion pieces to reflect a broad spectrum of views.
The article says several legal and regulatory developments are unfolding simultaneously. One example cited is a public interest litigation filed before India’s Supreme Court by journalist Geeta Seshu and the Software Freedom Law Centre, India, challenging parts of the DPDPA on constitutional and rights-related grounds. According to the piece, the Supreme Court later issued a notice to the Government of India on 12 March.
Concerns outlined in the article include the absence of journalistic exemptions, the lack of compensation for data breach victims when penalties are imposed to the government, broad state powers to exempt departments from the law, and questions about the independence of the Data Protection Board given the government’s control over appointments. The article notes that similar petitions had already been filed, but says this was the first time the court issued notice to the government.
The article also turns to proceedings before the Kerala High Court involving privacy concerns about biometric and personal data collected through Digi Yatra, a not-for-profit foundation that operates airport passenger-processing infrastructure in India. According to the piece, a public interest litigation filed by C R Neelakandan asked for a temporary restraint on the sharing of collected personal data and its commercial use without proper authorisation.
The article says the Kerala High Court issued notice to the Digi Yatra Foundation and sought clarification from the government on whether the Data Protection Board had been established to oversee such matters.
Alongside the litigation, the opinion piece points to government efforts to show legal preparedness for AI-related risks. It says Electronics and Information Technology Minister Ashwini Vaishnaw outlined existing safeguards during the ongoing parliamentary session, referring to the Information Technology Act, the DPDPA, and subordinate rules, along with published guidelines on AI governance, toy safety, harmful content, awareness-building measures, and cyber safety.
Cybersecurity developments also feature in the article. It says the Indian Computer Emergency Response Team, working with the SatCom Industry Association, issued guidelines on 26 February for space, including satellite communications. According to the piece, the framework is intended to strengthen resilience in India’s space ecosystem.
It applies to covered entities, including government agencies, satellite service providers, ground station operators, terminal equipment vendors, and private space entities. Incident reporting within six hours and annual audits are among the measures described.
A further section of the article draws on Thales’ 2026 Data Threat Report. The piece says 64% of surveyed organisations in India identified AI-driven transformation as their biggest security risk, while 55% said they had to deal with reputational damage caused by AI-generated misinformation. It also says 65% reported deepfake-driven attacks, 35% had a complete view of their data, and 36% could fully classify their data.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
OpenAI is moving to shut down the Sora app, its consumer-facing AI video platform, according to an official X post on 24 March. The move follows months of scrutiny around AI-generated video, including concerns over deepfakes, copyright, and harmful synthetic media.
The reported shutdown comes shortly after OpenAI retired Sora 1 in the United States on 13 March 2026 and replaced it with Sora 2 as the default experience. OpenAI’s help documentation says the older version remains available only in countries where the newer one has not yet launched, while support pages for the standalone Sora app are still live. The product changes also follow the announcement of new copyright settings for the latest video generation model.
That makes the current picture more complex than a simple sunset. Public OpenAI help pages still describe tools on iOS, Android, and the web, while news reports say the company has now decided to wind down the app itself. OpenAI had also recently indicated that it plans to integrate Sora video generation into ChatGPT, which could help explain why the standalone product is being reconsidered.
Sora became one of OpenAI’s most visible consumer media products, but it also drew sustained scrutiny over deepfakes, non-consensual content, and copyrighted characters. Such concerns remained central even as OpenAI added additional controls to the platform, including new consent and traceability measures to enhance AI video safety. AP reported that pressure from advocacy groups, scholars, and entertainment-sector voices formed part of the backdrop to the shutdown decision.
For users, the immediate issue is preservation of existing content. OpenAI’s Sora 1 sunset FAQ says some legacy material may be exportable for a limited period before deletion, but the company has not yet published a detailed standalone help document explaining the full shutdown. Based on the information now available, the clearest distinction is that OpenAI first retired one legacy version in some markets and is now reportedly ending the standalone app more broadly.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Over the past few years, we have witnessed a rapid shift in the way data is stored and processed across businesses, organisations, and digital systems.
What we are increasingly seeing is that AI itself is changing form as computation shifts away from centralised cloud environments to the network edge. Such a shift has come to be known as edge AI.
Edge AI refers to the deployment of machine learning models directly on local devices such as smartphones, sensors, industrial machines, and autonomous systems.
Instead of transmitting data to remote servers for processing, analysis is performed on the device itself, enabling faster responses and greater control over sensitive information.
Such a transition marks a significant departure from earlier models of AI deployment, where cloud infrastructure dominated both processing and storage.
From centralised AI to edge intelligence
Traditional AI systems used to rely heavily on centralised architectures. Data collected from users or devices would be transmitted to large-scale data centres, where powerful servers would perform computations and generate outputs.
Such a model offered efficiency, scalability, and easier security management, as protection efforts could be concentrated within controlled environments.
Centralisation allowed organisations to enforce uniform security policies, deploy updates rapidly, and monitor threats from a single vantage point. However, reliance on cloud infrastructure also introduced latency, bandwidth constraints, and increased exposure of sensitive data during transmission.
Edge AI introduces a fundamentally different paradigm. Moving computation closer to the data source reduces the reliance on continuous connectivity and enables real-time decision-making.
Such decentralisation represents not merely a technical shift but a reconfiguration of the way digital systems operate and interact with their environments.
Advantages of edge AI
Reduced latency and real-time processing
Latency is significantly reduced when computation occurs locally. Edge systems are particularly valuable in time-sensitive applications such as autonomous vehicles, healthcare monitoring, and industrial automation, where delays can have critical consequences.
Enhanced privacy and data control
Privacy improves when sensitive data remains on-device instead of being transmitted across networks. Such an approach aligns with growing concerns around data protection, regulatory compliance, and user trust.
Operational resilience
Edge systems can continue functioning even when network connectivity is limited or unavailable. In remote environments or critical infrastructure, independence from central servers ensures service continuity.
Bandwidth efficiency and cost reduction
Bandwidth consumption is decreased because only processed insights are transmitted, not raw data. Such efficiency can translate into reduced operational costs and improved system performance.
Personalisation and context awareness
Devices can adapt to user behaviour in real time, learning from local data without exposing sensitive information externally. In healthcare, personalised diagnostics can be performed directly on wearable devices, while in manufacturing, predictive maintenance can occur on-site.
The dark side of edge AI
However, the shift towards edge computing introduces profound cybersecurity challenges. The most significant of these is the expansion of the attack surface.
Instead of a limited number of well-protected data centres, organisations must secure vast networks of distributed devices. Each endpoint represents a potential entry point for malicious actors.
The scale and diversity of edge deployments complicate efforts to maintain consistent security standards. Security is no longer centralised but dispersed, increasing the likelihood of vulnerabilities and misconfigurations.
Let’s take a closer look at some other challenges of edge AI.
Physical vulnerabilities and device exposure
Edge devices often operate in uncontrolled environments, making physical access a major risk. Attackers may tamper with hardware, extract sensitive information, or reverse engineer AI models.
Model extraction attacks allow adversaries to replicate proprietary algorithms, undermining intellectual property and enabling further exploitation. Such risks are significantly more pronounced compared to cloud systems, where physical access is tightly controlled.
Software constraints and patch management challenges
Many edge devices rely on embedded systems with limited computational resources. Such constraints make it difficult to implement robust security measures, including advanced encryption and intrusion detection.
Patch management becomes increasingly complex in decentralised environments. Ensuring that millions of devices receive timely updates is a significant challenge, particularly when connectivity is inconsistent or when devices operate in remote locations.
Breakdown of traditional security models
The decentralised nature of edge AI undermines conventional perimeter-based security frameworks. Without a clearly defined boundary, traditional approaches to network defence lose effectiveness.
Each device must be treated as an independent security domain, requiring authentication, authorisation, and continuous monitoring. Identity management becomes more complex as the number of devices grows, increasing the risk of misconfiguration and unauthorised access.
Data integrity and adversarial threats
As we mentioned before, edge devices rely heavily on local data inputs to make decisions. As a result, manipulated inputs can lead to compromised outcomes. Adversarial attacks, in which inputs are deliberately altered to deceive machine learning models, represent a significant threat.
In safety-critical systems, such manipulation can lead to severe consequences. Altered sensor data in industrial environments may disrupt operations, while compromised vision systems in autonomous vehicles may produce dangerous behaviour.
Supply chain risks in edge AI
Edge AI systems depend on a combination of hardware, software, and pre-trained models sourced from multiple vendors. Each component introduces potential vulnerabilities.
Attackers may compromise supply chains by inserting backdoors during manufacturing, distributing malicious updates, or exploiting third-party software dependencies. The global nature of technology supply chains complicates efforts to ensure trust and accountability.
Energy constraints and security trade-offs
Edge devices are often designed with efficiency in mind, prioritising performance and power consumption. Security mechanisms such as encryption and continuous monitoring require computational resources that may be limited.
As a result, security features may be simplified or omitted, increasing exposure to cyber threats. Balancing efficiency with robust protection remains a persistent challenge.
Cyber-physical risks and real-world impact
The integration of edge AI into cyber-physical systems elevates the consequences of security breaches. Digital manipulation can directly influence physical outcomes, affecting safety and infrastructure.
Compromised healthcare devices may produce incorrect diagnoses, while disrupted transportation systems may lead to accidents. In energy networks, attacks could impact entire regions, highlighting the broader societal implications of edge AI vulnerabilities.
Regulatory and governance challenges
Existing regulatory frameworks have been largely designed for centralised systems and do not fully address the complexities of decentralised architectures. Questions regarding liability, accountability, and enforcement remain unresolved.
Organisations may struggle to implement effective security practices without clear standards. Policymakers face the challenge of developing regulations that reflect the distributed nature of edge AI systems.
Towards a secure edge AI ecosystem
Addressing all these challenges requires a multi-layered and adaptive approach that reflects the complexity of edge AI environments.
Hardware-level protections, such as secure enclaves and trusted execution environments, play a critical role in safeguarding sensitive operations from physical tampering and low-level attacks.
Encryption and secure boot processes further strengthen device integrity, ensuring that both data and models remain protected and that unauthorised modifications are prevented from the outset.
At the software level, continuous monitoring and anomaly detection are essential for identifying threats in real time, particularly in distributed systems where central oversight is limited.
Secure update mechanisms must also be prioritised, ensuring that patches and security improvements can be deployed efficiently and reliably across large networks of devices, even in conditions of intermittent connectivity.
Without such mechanisms, vulnerabilities can persist and spread across the ecosystem.
Rather than relying entirely on decentralised or centralised models, organisations are distributing workloads strategically, keeping latency-sensitive and privacy-critical processes on the edge while maintaining centralised oversight, analytics, and security coordination in the cloud.
Such an approach allows organisations to balance performance and control, while enabling more effective threat detection and response through aggregated intelligence.
Security must also be embedded into system design from the outset, rather than treated as an additional layer to be applied after deployment. A proactive approach to risk assessment, combined with secure development practices, can significantly reduce vulnerabilities before systems are operational.
In conclusion, we have seen how the rise of edge AI represents a pivotal shift in both AI and cybersecurity. Decentralisation enables faster, more private, and more resilient systems, yet it also creates a fragmented and dynamic attack surface.
The advantages we have outlined are compelling, but they also introduce additional layers of complexity and risk. Addressing these challenges requires a comprehensive approach that combines technological innovation, regulatory development, and organisational awareness.
Only through such coordinated efforts can the benefits of edge AI be realised while ensuring that security, trust, and safety remain intact in an increasingly decentralised digital landscape.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
