Telecom’s AI Shift Moves From Chatbots to Networks That Can Diagnose and Repair Themselves

Artificial intelligence in telecommunications is moving beyond customer-service chatbots and experimental projects towards a much larger ambition: networks capable of identifying problems, predicting failures and eventually taking corrective action themselves. At Ai4 2026 in Las Vegas, a panel examining how AI is transforming telecom brought together perspectives spanning network infrastructure, machine learning, customer experience and autonomous-network research. Moderated by Cathy White, founder and CEO of CEW Communications, the discussion included Sahil Yadav, Head of Software and AI at Applied Optoelectronics; Sundar Krishnan, Director of AI and Machine Learning at Optimum; and telecom veteran and Bell Labs Fellow Anne Lee.

The discussion suggested that telecommunications may become one of the industries where the transition from generative AI to agentic AI has particularly visible consequences. Rather than simply helping employees generate information, AI systems are beginning to interact with the physical and digital infrastructure responsible for keeping millions of customers connected. For telecom operators, some of the clearest opportunities are already emerging in network operations. Predictive maintenance, anomaly detection and automated diagnostics allow operators to identify potential failures before customers necessarily experience them. The next step is self-healing, where AI not only identifies the problem but recommends or ultimately executes the solution.

Yadav described this transition from the perspective of network infrastructure. Applied Optoelectronics supplies communications equipment and associated software, and its AI work includes systems intended to support network optimisation and increasingly autonomous operations. The potential economic impact is considerable. Network outages create direct repair costs while also affecting customer satisfaction, retention and an operator’s reputation. AI capable of identifying deteriorating equipment before failure could allow maintenance teams to intervene earlier, while better diagnostics could reduce unnecessary field visits.

However, fully autonomous telecommunications networks remain some distance away. The panel drew an important distinction between specialised automation and an entire network capable of continuously managing and evolving itself. Lee, whose career has covered successive generations of telecommunications technology, argued that the nearer-term opportunity is likely to involve AI agents designed to handle specific network problems. An agent could recognise a recurring failure pattern, determine an appropriate response and potentially resolve it automatically within clearly defined limits.

Over time, larger numbers of specialised agents could work together across different parts of the network. This is consistent with research into self-operating networks, where agentic AI, digital twins and other technologies could eventually support infrastructure capable of adapting dynamically with substantially less manual intervention. The progression will probably be gradual because the consequences of an incorrect decision can be significant. Yadav described an approach based on risk. Low-risk and easily reversible actions can increasingly be assigned to AI. Medium-risk actions can operate under human supervision, while decisions capable of causing significant network disruption should continue to require human approval.

This could become one of the most important principles governing AI deployment in critical infrastructure. The objective is not necessarily to make every network function autonomous as quickly as possible. It is to expand autonomy as confidence, testing and operational experience improve. AI is also changing the economics of field operations. Telecom providers routinely dispatch technicians to investigate customer and network problems, but sending a vehicle and engineer to every suspected fault is expensive. Better prediction and remote diagnosis can help operators determine which problems genuinely require physical intervention.

Customer applications can increasingly perform initial diagnostics themselves, while AI can guide subscribers through basic troubleshooting. Visual AI could eventually allow customers to photograph equipment and receive instructions based on what the system sees. More sophisticated network models can analyse signal quality, connected devices and historical faults to determine whether a problem originates inside a home, elsewhere in the local network or within wider infrastructure. This creates a direct connection between AI investment and operating expenditure. The value is not the number of AI queries generated but whether operators reduce outages, unnecessary service calls, customer churn and the amount of engineering time spent on routine problems.

The customer side of telecommunications is undergoing a parallel transformation. Krishnan described how recommendation systems have progressed from relatively simple rules towards machine-learning and AI systems capable of providing customer-service representatives with more contextual guidance. Instead of waiting for a subscriber to explain the entire history of a problem, an AI system can potentially analyse the customer’s relationship with the operator, recent service issues, products and previous interactions before the conversation begins.

That could allow the representative to understand whether someone is calling because they want to upgrade a service, complain about unreliable broadband or potentially cancel their subscription. The system can then suggest relevant information, offers or potential resolutions. AI could also make customer retention more proactive. Rather than waiting until a dissatisfied subscriber contacts the operator to cancel, predictive systems can identify behavioural patterns associated with potential churn and allow the company to intervene earlier.

Digital twins could extend this further. Krishnan described experiments involving simulated customers that allow companies to test how different types of subscribers might respond to offers or service decisions before those approaches are deployed more widely. The underlying objective is hyper-personalisation, but the commercial test remains straightforward: does the technology improve retention, service quality or revenue sufficiently to justify its cost?

That question is becoming more important as corporate attitudes towards AI expenditure mature. The first phase of the generative-AI boom encouraged companies to experiment widely, often measuring adoption through the number of users, prompts or tokens consumed. The next phase is increasingly about linking expenditure to business results. For telecom companies, that means distinguishing between an impressive demonstration and a system capable of operating reliably across millions of subscribers or network devices.

Moving from a proof of concept to production requires considerably more than a functioning AI model. Companies need data pipelines, access controls, monitoring, testing, governance, latency management, security and people capable of maintaining the resulting system. This helps explain why promising AI pilots do not automatically become commercial deployments. Building a demonstration may prove that a technology works. Scaling it requires proving that it remains reliable when exposed to the complexity of a real telecommunications network.

The industry has faced similar challenges before. Each major generation of telecommunications technology has moved through prototypes and limited deployments before reaching commercial scale. AI is different technologically, but many of the engineering disciplines required for deployment remain familiar: reliability, capacity, security and operational resilience.

Legacy infrastructure presents another major obstacle. AI depends heavily on data, but large telecommunications networks contain equipment installed across many different technological generations. Some devices can provide detailed real-time telemetry and be controlled remotely, while older infrastructure may provide relatively little information. Making networks more intelligent therefore has a physical investment dimension.

Operators cannot simply install an AI platform and expect decades-old network equipment to become autonomous. In some cases, equipment needs additional sensing and telemetry capabilities. In others, older devices may ultimately need replacement. That makes AI deployment partly a capital-expenditure question. Operators need to determine where upgrading infrastructure creates enough operational value to justify the investment.

The panel suggested that phased deployment offers one practical solution. An operator can modernise equipment within a particular geographic area, introduce AI capabilities there and observe how the technology performs before extending it across the wider network. This approach also limits the potential impact of failure. If an autonomous system behaves incorrectly, the affected portion of the network remains relatively contained while engineers gain real-world experience that can improve subsequent deployments.

Edge AI could accelerate this transformation further. Smaller and more efficient models increasingly allow AI processing to take place closer to the network device or customer rather than sending every request to a large central cloud environment. For telecom companies, edge processing can offer advantages including lower latency, greater resilience and potentially better control of sensitive information. Specialised AI agents could perform relatively narrow functions on network equipment without requiring the computing resources associated with the largest generative models.

The relationship between AI and telecommunications consequently works in both directions. Telecom companies can use AI to operate their infrastructure more efficiently, but the growth of AI itself is also increasing demand for faster, more responsive and more distributed communications infrastructure. Industry research in 2026 increasingly describes this as a shift from using AI within telecom towards telecom becoming part of the infrastructure enabling the wider AI economy.

The human consequences may be equally significant. Network engineers are unlikely simply to disappear as automation increases. Their responsibilities are more likely to change. An engineer who previously spent much of the working day identifying and correcting routine network faults may increasingly supervise automated systems, investigate exceptional situations and manage the AI agents operating parts of the infrastructure. Understanding machine-learning systems, evaluating AI recommendations and determining when automation should be overridden could consequently become important telecom engineering skills.

Lee argued that releasing engineers from repetitive network-management work could also create capacity for new services. Telecom companies have spent enormous resources maintaining increasingly complicated networks. Greater automation could allow some of those resources to move towards developing products that generate additional revenue or provide wider social value. Emergency communications provide one example. Modern networks are technically capable of transmitting far more than voice, including video, images and other information that could potentially assist emergency responders. Developing and operating such services requires investment, however, and efficiency gains elsewhere in the network could make these types of applications more economically practical.

Customer service could experience an equally visible transformation. Traditional interactive voice-response systems remain a frequent source of frustration because users navigate layers of menus before reaching the information or person they need. More capable conversational AI could eventually replace much of that structure. Instead of selecting options through a telephone keypad, customers could describe the problem naturally and allow the system to determine what information or action is required.

The important threshold will be whether customers stop wanting to escape from the automated system and speak immediately to a human. Reaching that point will require better contextual understanding, reliability and trust rather than simply more convincing synthetic voices.

Telecommunications therefore provides a useful illustration of where enterprise AI is heading. The industry is moving beyond the question of whether companies should experiment with AI. The more important questions now concern which functions should be automated, how quickly autonomy should expand and whether the economic return justifies the infrastructure required to support it.

The ultimate destination may be networks capable of continuously monitoring their own performance, anticipating failures, reallocating resources and resolving many problems without waiting for engineers to intervene. Getting there will require modernised physical infrastructure, reliable data, governance, human oversight and a gradual transfer of responsibility from engineers to machines.

The result would represent more than another efficiency programme. Telecommunications networks could evolve from infrastructure that humans continuously operate into infrastructure that increasingly operates itself, with engineers supervising the intelligence responsible for keeping it running.

UK authorities intensify action against unregistered crypto trading businesses

UK financial and law-enforcement authorities have carried out another operation against businesses suspected of providing unregistered peer-to-peer cryptocurrency trading services in London, extending an enforcement campaign that began earlier this year.

The Financial Conduct Authority worked with HM Revenue & Customs and the Metropolitan Police on the operation, which took place on 10 September and covered three premises in the capital. The businesses were instructed to stop the activities under investigation through cease-and-desist notices issued at each location.

The action focuses on people conducting direct cryptocurrency transactions as a commercial activity rather than individuals occasionally buying or selling digital assets privately. Businesses providing this type of service in the UK are required to meet the applicable registration requirements. The FCA says there are currently no peer-to-peer crypto businesses registered with it.

Regulators are particularly concerned that businesses operating outside the registration system can provide channels through which illegally obtained money is transferred or concealed. Registration under the UK’s anti-money-laundering framework requires businesses to maintain controls intended to identify suspicious activity and reduce financial-crime risks.

The latest intervention follows an earlier operation in April, when the FCA and partner authorities targeted eight London premises. Information collected during that operation is now contributing to criminal investigations and other enforcement activity.

The FCA has been increasing its activity around unregistered digital-asset businesses as the UK prepares to introduce a substantially broader regulatory framework for crypto. Previous enforcement has included the prosecution of an operator of an unlawful cryptocurrency ATM network and cooperation with police investigations into suspected illegal crypto exchanges.

The timing is significant because the regulatory environment for the sector is entering a transitional period. The FCA has provided further guidance explaining which activities will fall within the UK’s future crypto regulatory system, with the broader framework scheduled to take effect in October 2027.

The expanded regime will bring a wider range of digital-asset activities within financial regulation. Businesses already operating in the sector will therefore need to determine whether their activities require authorisation under the new framework rather than assuming that an existing registration will be sufficient.

Until the wider framework takes effect, much of the UK crypto market remains outside the regulatory system applying to conventional financial services, although requirements covering areas including money laundering and financial promotions already apply. The September enforcement operation was conducted under the UK’s existing anti-money-laundering legislation.

The latest London action therefore comes as the UK moves from a relatively limited regulatory approach towards a more extensive authorisation system for digital assets. For businesses operating in the market, the combination of enforcement against unregistered activity and preparations for the 2027 regime points to increasing regulatory scrutiny before the new framework formally takes effect.

Source: CMS

China sets nationwide court framework for disputes involving artificial intelligence

China’s Supreme People’s Court has issued national guidance for handling disputes involving artificial intelligence, giving businesses, technology developers and users greater clarity over how existing Chinese law will be applied to issues ranging from copyright and personal data to deepfakes, consumer protection and autonomous vehicles.

Published on 7 September 2026, the guidance contains 24 provisions divided into five sections. It does not create a separate AI statute. Instead, it directs courts on how existing legislation, including China’s Civil Code, Copyright Law, Personal Information Protection Law, Cybersecurity Law, Data Security Law and consumer protection rules, should be applied when AI is involved in a dispute.

One of the central principles concerns responsibility when AI causes harm. Unless another law establishes a different standard, courts are instructed to examine whether the party involved was at fault. Judges can consider how an AI system was being used, its degree of autonomy and transparency, the potential scale of harm, safeguards introduced by developers or service providers and the extent to which users could anticipate or control the consequences.

The approach is particularly relevant to generative AI businesses because responsibility will not automatically fall on one participant in the technology chain. Developers, providers and users can each face scrutiny depending on their role and conduct. Service providers that become aware of content infringing personal rights may also face consequences if they fail to take appropriate action.

Intellectual property is another major part of the framework. In copyright disputes involving AI-generated material, courts are instructed to consider factors including the type of service, the origins of training material, the involvement of different parties, preventative measures and any financial benefit obtained. Developers arguing that their systems have not infringed copyright may be required to provide evidence concerning training-data sources, training records, how the model operates and the technical reasoning supporting their position.

Users can also be held responsible. Where someone knew or should have known about an existing protected work and uses AI to produce substantially similar material without an adequate legal justification, a court may find in favour of the rights holder. The framework therefore places potential responsibility on both the companies building AI systems and the people using them.

However, the Supreme People’s Court has deliberately avoided resolving some of the most contested questions surrounding generative AI. The guidance does not provide a comprehensive answer to every issue concerning copyright protection for AI output or the use of copyrighted material in model training. Areas where sufficient agreement has not yet developed remain open to further interpretation through future cases.

Personal identity and privacy receive substantial attention. Creating and distributing an identifiable digital representation of an individual without permission can infringe personal rights. Similar considerations apply to synthetic voices, while AI-generated material that damages the reputation of an identifiable individual can also create potential liability.

The court has also addressed the use of information that is already publicly available. Public accessibility does not automatically make personal information unrestricted material for AI development. Whether information can be used for model training can depend on the purpose, necessity and scale of the processing, the sensitivity of the information and its potential effect on the individual concerned.

Consumer-facing applications are another focus. The guidance addresses the use of algorithms to impose unjustified differences in prices or other transaction conditions on consumers. It also covers fraudulent AI impersonation, including situations where synthetic identities or representations of well-known individuals are used to deceive consumers.

For companies developing AI-enabled physical products, courts can consider factors such as intended use, self-learning capabilities, software updates, the degree of control available to users, relevant technical standards and whether manufacturers adequately disclosed foreseeable limitations and risks when assessing potential product defects.

Automotive businesses are specifically covered. In disputes involving automated or driver-assistance systems, courts may require parties controlling relevant information to provide authentic and complete event records where these are necessary to determine the circumstances surrounding an accident.

Open-source development is also addressed. Courts considering disputes involving open-source AI components can examine licence conditions, restrictions on use, security precautions and information supplied by developers. This provides a basis for distinguishing the responsibilities of original developers from businesses that subsequently incorporate, distribute or modify open-source components.

The guidance also deals with AI-related inventions. Technical innovations capable of patent protection remain subject to existing intellectual-property rules, while recognition as an inventor depends on a natural person having made a creative contribution to the substantive features of the invention.

AI use in legal proceedings is subject to additional safeguards. Parties using AI to prepare pleadings, legal research or other materials submitted to courts are expected to verify their accuracy, disclose the use of AI assistance and remain responsible for the information submitted. Courts can also examine the authenticity and integrity of electronic and AI-generated evidence.

For international companies operating AI systems in China, the practical consequences extend beyond legal interpretation. The framework increases the importance of maintaining records concerning training-data sources, model development, technical safeguards, open-source licences and other decisions affecting how AI systems operate. Such information could become important evidence if a dispute reaches a Chinese court.

Rather than introducing a separate layer of AI legislation, the Supreme People’s Court has established a common judicial approach for applying China’s existing laws to rapidly developing technology. For developers, technology providers and multinational businesses, the guidance provides a clearer indication of how responsibility may be assessed, while several of the most difficult questions surrounding generative AI and intellectual property remain open for future court decisions.

Source: CMS

Netherlands targets rental housing investment with lower property transaction tax

The Dutch government has proposed a series of tax changes aimed at improving investment conditions in the housing market, including a reduction in the transaction tax applying to residential properties purchased by investors and additional financial support for housing corporations.

The measures form part of the 2027 Tax Plan submitted to the House of Representatives on 15 September. They remain subject to parliamentary approval and could be amended before taking effect. Most measures in the package are intended to apply from 1 January 2027.

For residential property investors, the most direct change is a proposed reduction in real estate transfer tax from 8% to 7% for homes that will not be occupied by their purchaser. The measure covers properties acquired for rental purposes as well as other residential purchases where the buyer does not establish their own residence. The existing 2% rate for qualifying owner-occupiers is unaffected.

Reducing the rate by one percentage point lowers the upfront cost of acquiring rental housing and also reduces the tax burden when investment properties are transferred between landlords. The government has linked the measure to its efforts to create more favourable conditions for private investment in the rental sector.

The change comes as the Netherlands continues to face a significant shortage of housing and the government seeks to increase both private and public-sector development capacity. Alongside the tax proposals, approximately €7 billion has been allocated to housing measures through 2035, including €2.3 billion for incentives connected with affordable housing construction. Municipalities are expected to receive €7,000 for each qualifying affordable home delivered under this programme.

A further €940 million has been allocated for major housing development locations, while €650 million is intended to support projects that otherwise face difficulties reaching financial viability. Another €180 million is being made available to help municipalities acquire land earlier in the development process.

Private rental development is also expected to receive €500 million through support directed at middle-income rental housing. Combined with the lower transfer tax, the measures are intended to improve the economics of projects and acquisitions at a time when financing costs, construction expenses and regulation continue to influence residential investment decisions.

Housing corporations are another major focus of the tax package. From 2027, qualifying transfers of social housing between housing corporations would become exempt from transfer tax. The change is designed to make it easier for organisations to redistribute housing portfolios without generating an additional transaction-tax cost.

A potentially larger financial change is planned from 2028. The government proposes removing the general restriction on interest deductions for housing corporations. The measure is intended to increase the amount of capital available for new social housing and improvements to existing properties. Together with other housing measures, the government estimates that its programme could provide corporations with around €28 billion of additional investment capacity.

The 2027 package also proposes increasing the deduction available for qualifying energy investments from 40% to 45.5% from next year, providing another potential incentive for businesses undertaking eligible energy-related expenditure.

Taken together, the proposals show the government using taxation alongside direct public funding to address constraints on Dutch housing delivery. Private investors would face a lower acquisition cost for rental homes, while housing corporations would receive tax changes intended to strengthen their capacity to build, acquire and renovate residential stock.

The measures are not yet final. The 2027 Tax Plan and related legislation must pass through the Dutch Parliament, and individual provisions could still be amended before taking effect.

Source: CMS

Swedish court sets clearer boundary for public-sector leases involving new construction

A ruling by Sweden’s Supreme Administrative Court has provided new guidance for property owners and public-sector occupiers entering lease agreements that combine existing premises with additional space to be constructed by the landlord.

The case concerned Nacka Municipality and school premises it had occupied for several years. Faced with the need for additional capacity, the municipality entered a new ten-year agreement with its existing landlord, Sickla Industrifastigheter KB, without conducting a public procurement procedure. The agreement covered approximately 8,155 sqm in the existing building together with around 2,120 sqm in an adjoining extension that the landlord would develop.

The Swedish Competition Authority challenged the arrangement, arguing that the construction component meant the agreement should have been competitively procured. The dispute eventually reached the Supreme Administrative Court, which issued its judgment on 11 September 2026.

An important part of the case concerned the planned extension. The municipality had influenced its design and requested modifications specifically suited to the school’s requirements. The court therefore concluded that the construction of the additional premises had the characteristics of a public works contract when considered separately.

That finding did not determine the treatment of the entire agreement. Because the contract combined the continued occupation of an existing property with the construction of additional premises, the court considered which element represented the main purpose of the overall transaction.

The municipality was already operating the school at the property and required additional capacity to accommodate more pupils. Discussions with the landlord resulted in plans to use development potential immediately beside the existing school rather than establishing an entirely new facility elsewhere.

After considering the arrangement as a whole, the court concluded that its main purpose was to allow the municipality to continue occupying and operating the existing school. The additional building was considered supplementary to that objective from both financial and operational perspectives.

The overall agreement was consequently treated as a property lease falling outside the procurement requirements that would otherwise apply to a public works contract. Nacka Municipality was therefore not required to conduct a competitive procurement process for the agreement, and there was no basis for imposing the procurement penalty sought in the case.

The decision has wider implications for Swedish commercial property because municipalities and other public bodies are significant occupiers of privately owned buildings. Landlords can be required to refurbish, extend or adapt properties when negotiating leases with public-sector tenants, creating uncertainty over when property arrangements cross into procurement-regulated construction.

The ruling indicates that the inclusion of substantial construction work does not automatically determine the legal treatment of the entire property agreement. Instead, the purpose and structure of the transaction have to be considered as a whole, including the relationship between the existing premises and any additional development.

At the same time, the judgment does not establish a fixed percentage of new construction that can automatically be incorporated into an exempt property lease. In the Nacka case, approximately 2,120 sqm was being added alongside 8,155 sqm of existing premises, but the court focused on the function of the extension within the wider agreement rather than creating a general floor-area threshold.

This distinction is particularly relevant for developers considering projects involving public-sector tenants. An extension supporting an organisation already occupying a property may be assessed differently from a purpose-built development where the public body substantially determines the design and delivery of a new building.

The judgment therefore provides additional guidance for extensions and adaptations associated with existing public-sector leases without creating a general exemption for landlord-led development. Larger build-to-suit projects and agreements primarily concerned with delivering new premises will still require individual assessment to establish whether public procurement rules apply.

For property owners and developers, the decision makes the underlying purpose of the agreement particularly important. Where construction supports the continuation of an established tenancy, it may remain secondary to the property lease. Where delivering a new building is effectively the central objective, the legal position may be different, leaving the boundary between a property lease and a procurement-regulated construction contract dependent on the circumstances of each project.

Source: CMS

CNB holds rates at 3.75% as financing pressure persists across Czech property market

The Czech National Bank kept its main interest rate at 3.75% on 17 September, extending the period of tighter financing conditions for households, developers and property investors while leaving open the possibility of another increase later this year.

All seven members of the Bank Board supported maintaining the two-week repo rate at its current level. The decision follows an unchanged rate in August and a 25-basis-point increase in June, when the CNB lifted the rate from 3.50% to 3.75%. The discount rate remains at 2.75% and the Lombard rate at 4.75%.

The central bank considers the June increase to have tightened monetary conditions sufficiently for the moment. Longer-term interest rates have also moved higher, adding further restraint to financing conditions and potentially slowing the expansion of new lending. At the same time, the CNB sees greater inflation risk coming from commodity markets, while underlying domestic price pressures remain a concern.

Inflation has remained close to the central bank’s 2% target since the beginning of 2024, but the CNB expects a temporary acceleration around the end of 2026 and beginning of 2027. Core inflation remains elevated, leaving policymakers cautious about declaring the inflation problem resolved.

Domestic economic conditions are adding to that caution. Wage growth remains relatively strong against a tight labour market, while faster credit expansion and debt-financed public expenditure could contribute to additional price pressure. At the same time, economic growth has moderated. Czech GDP growth slowed from 2.2% year-on-year in the first quarter to 1.9% in the second quarter of 2026, with domestic demand continuing to provide the main support to the economy.

For the Czech property sector, the decision means there is no immediate relief through lower central bank rates. Developers and investors continue to face relatively restrictive financing conditions, making the price and availability of debt important considerations for acquisitions, refinancing and new construction.

Residential property is particularly exposed to the interest-rate environment. Mortgage costs influence both monthly repayments and the amount households can borrow, meaning financing conditions remain an important constraint on affordability even where underlying demand for housing is strong. Property-price growth is also among the developments being watched by the central bank as it assesses inflation and financial conditions.

Commercial real estate is affected through investment and development financing. The cost of debt influences acquisition returns and project feasibility, while uncertainty over the future direction of rates can make it more difficult for buyers and sellers to agree on pricing. Refinancing remains another important consideration for owners whose existing debt is approaching maturity.

The September decision should therefore not be interpreted as the beginning of another easing cycle. The CNB has indicated that its next decision could involve either maintaining the current rate or increasing it, depending on incoming economic and inflation data. The next scheduled monetary policy meeting is on 5 November.

For Czech real estate, the immediate picture is consequently one of continued financing discipline rather than cheaper capital. The 3.75% rate provides stability for now, but the possibility of renewed tightening means developers, investors and homebuyers cannot yet assume that borrowing costs are heading lower.

Source: CTK

Ericsson relocation deepens contrast across Stockholm office market

Stockholm’s office market is showing an increasingly pronounced difference between its stronger urban districts and locations facing higher vacancy, with Ericsson’s planned relocation from Kista to Hagastaden set to become one of the clearest examples of this changing pattern.

During the second quarter of 2026, Ericsson announced agreements covering around 71,000 sqm of new office space with Atrium Ljungberg and Castellum in Hagastaden. Together with premises previously secured in the district, the agreements would increase the company’s total presence there to approximately 95,000 sqm.

The largest part of the expansion involves approximately 58,000 sqm across three planned buildings being developed by Atrium Ljungberg. The leases have 15-year terms, with occupation expected in stages between 2031 and 2033. The agreements are subject to conditions connected with land allocation and development agreements, which are expected to be resolved by the end of 2027.

The relocation will have substantial consequences for Kista, where Ericsson is expected eventually to vacate around 123,000 sqm. Based on vacancy conditions recorded in Q2 2026, CBRE estimates that Kista’s vacancy rate could reach between 55% and 60% once Ericsson has left. The estimate illustrates the scale of the challenge facing a district that already has considerably more available office space than Stockholm’s strongest locations.

Conditions in Stockholm’s CBD remain markedly different. Prime office rents were unchanged at SEK 9,500 per sqm annually during the second quarter, while the highest rents reached SEK 12,000 per sqm. The prime yield remained at 3.9%, and CBD vacancy declined from 7.0% in the first quarter to 6.8% in Q2. Across the wider Stockholm office market, vacancy stood at 14%, demonstrating how the citywide figure conceals substantial differences between individual districts.

Other large leasing agreements during the quarter also concentrated demand in established urban locations. Fastpartner leased 15,000 sqm at Tullvaktsvägen in Gärdet to Magasin X for an AI and robotics hub, while H&M Studios agreed a 6,900 sqm expansion in the same former Nasdaq property. Skanska also leased 6,600 sqm in its Olivin building in Västra Kungsholmen to Telenor, which plans to establish its Swedish headquarters there.

The transactions indicate that substantial corporate requirements remain in the Stockholm market, but demand is not being distributed evenly. Large occupiers are committing to particular districts and buildings, while locations losing major tenants face the more difficult task of finding new uses or occupiers for significant amounts of existing space.

Investment activity also remained substantial during the second quarter. Swedish office properties attracted SEK 15.7 billion of investment, representing around 20% of total property transaction volume and making offices the country’s third-largest investment sector during the period. However, the quarterly figure was influenced by several large transactions rather than a widespread increase in deal activity.

One of the largest transactions involved Castellum’s sale of two Stockholm office properties in Hagastaden and Norrmalm to Alecta Fastigheter for approximately SEK 5 billion. Alecta was also active in Gothenburg, where it acquired two fully occupied office properties in Gårda from Technopolis comprising approximately 28,500 sqm.

Ericsson’s relocation is therefore significant beyond the size of the individual leases. Hagastaden is positioned to gain one of Sweden’s largest corporate office concentrations, while Kista could eventually see more than 120,000 sqm released as Ericsson consolidates its operations elsewhere.

Combined with relatively low CBD vacancy, major leases in other inner-city districts and continued institutional investment in well-located properties, the development points to a Stockholm office market increasingly defined by differences between locations rather than a uniform movement in demand. For owners and investors, the ability of individual districts and buildings to retain major occupiers is consequently becoming an increasingly important factor in future performance.

Nordic hotel demand strengthens while investment market remains selective

Nordic hotels recorded a strong summer tourism season in 2026, with international visitors supporting performance across several major cities, while the investment market continued to face a more difficult environment shaped by financing conditions and differences between buyer and seller expectations.

Hotel transactions across the Nordic region reached €751 million between January and August 2026, involving 22 deals above €5 million. On the surface, investment volume was 54% lower than during the corresponding period of 2025. However, the comparison is heavily affected by CapMan’s large Midstar portfolio acquisition last year. Excluding that transaction, this year’s investment volume was around 5% higher than in the comparable period.

Several sizeable transactions continued during a period that normally experiences slower activity. In Copenhagen, AP Ejendomme and CapMan acquired the Crowne Plaza from Niam, while the property occupied by 25hours Hotel Copenhagen had previously changed ownership in March. In Finland, Balder expanded its hotel holdings through the acquisition of Lapland Hotels Arena in Tampere for approximately €40 million after also acquiring Scandic Meilahti in Helsinki earlier in the year.

Pricing conditions differed between the Nordic capitals. CBRE placed the prime yield for leased hotels at 4.70% in Copenhagen, 4.75% in Stockholm, 5.25% in Oslo and 6.00% in Helsinki. Stockholm’s benchmark was 25 basis points lower than a year earlier, Copenhagen was down 10 basis points and Helsinki was unchanged, while Oslo increased by 25 basis points.

Despite the increase in the number of transactions compared with the previous year, liquidity remains constrained. CBRE reports that some potential deals are being delayed or abandoned because buyers and sellers have struggled to agree on pricing. Higher Nordic swap rates and geopolitical uncertainty have added further complexity to investment decisions.

The operating side of the market presented a more positive picture during the summer. Denmark recorded RevPAR growth in three of the four markets covered by the research between May and August. Copenhagen increased 6.6% year-on-year to DKK 1,280, Aalborg rose 6.9% to DKK 509 and Aarhus advanced 1.0% to DKK 731. Odense was the exception, declining 2.0% to DKK 591. Copenhagen’s occupancy increased from 87.4% to 88.2%, while its average daily room rate reached DKK 1,451.

Sweden delivered some of the strongest growth among the markets covered. Stockholm’s RevPAR increased 12.0% to SEK 1,234, followed by Malmö with growth of 7.8% to SEK 932, Gothenburg at 6.1% to SEK 1,109 and Uppsala at 2.2% to SEK 801. International guest nights in Sweden increased 14%, compared with 3% growth from domestic travellers. Stockholm occupancy reached 76.0%, Gothenburg 79.2% and Malmö 82.1%.

Norway produced a more mixed result despite exceptionally strong visitor numbers. July hotel guest nights exceeded four million for the first time since the country’s hotel statistics series began in 1986. Bergen was the strongest major Norwegian city covered by the report, with RevPAR increasing 10.2% to NOK 1,774 between May and August. Stavanger recorded growth of 4.2% to NOK 1,066, while Oslo declined 0.9%, Trondheim fell 7.7% and Tromsø was down 7.2%.

Finland moved in the opposite direction to much of the region. Helsinki RevPAR declined 2.8% to €76, Turku fell 1.5% to €89 and Tampere decreased 4.1% to €76. Rovaniemi was the exception, recording a 14.0% increase to €41. Its improvement came from higher occupancy, which increased from 39% to 45%, while average room rates remained broadly unchanged.

The figures point to an increasingly differentiated Nordic hotel market rather than a uniform regional recovery. International leisure demand and stronger room pricing are supporting several Scandinavian cities, particularly Stockholm, Copenhagen and Bergen, while Finland’s larger urban markets experienced weaker summer performance.

Improving hotel operations have also not yet translated into a broad acceleration of investment activity. The €751 million invested through August was spread across a relatively limited number of transactions, while CBRE describes buyers as selective and reports continuing difficulties in bringing some potential deals to completion. This leaves location, operating performance and the potential to improve individual assets central to investment decisions across the Nordic hotel sector.

DL Invest secures €28.8m financing for Silesian retail park portfolio

DL Invest Group has secured €28.8 million in financing from Austrian lender HYPO NOE for a portfolio of 11 retail parks in southern Poland, providing the group with five-year funding against approximately 44,300 sqm of retail space.

The financing agreement was concluded on 25 August 2026, with the funds drawn during the first week of September. The transaction covers assets located entirely within the Silesian Voivodeship, one of DL Invest Group’s core operating markets.

The financed portfolio includes two retail parks in Siemianowice Śląskie and two in Knurów, alongside properties in Czeladź, Rybnik, Zawiercie, Mikołów, Katowice, Ruda Śląska and Rydułtowy. The locations primarily serve local and everyday shopping demand across the Upper Silesian-Zagłębie metropolitan area and surrounding cities.

The transaction adds an international banking relationship to DL Invest Group’s funding structure as the company continues developing and managing its Polish commercial property portfolio. Rather than pursuing asset disposals as its principal strategy, the group has focused on retaining properties and managing them over the longer term.

“Securing five-year financing from HYPO NOE confirms the quality of our retail park portfolio, the stability of the cash flows it generates and DL Invest Group’s credibility as a long-term partner for international financial institutions,” said Dominik Leszczyński, Founder and CEO of DL Invest Group.

The financing follows a period of expansion for the Katowice-based property company. DL Invest Group reported revenue of PLN 206.1 million for the first half of 2026, an increase of 35% compared with the corresponding period of 2025, while EBITDA reached PLN 115.8 million, up 52.4%. The group has reported assets of approximately €1.2 billion and portfolio occupancy of around 97%.

Retail parks form part of DL Invest Group’s wider portfolio strategy, alongside other commercial property sectors. The latest financing demonstrates continued lender appetite for established regional retail assets capable of generating recurring rental income, while giving DL Invest additional medium-term funding as it continues expanding its property platform in Poland.

REWE Group secures CFO leadership through 2032

REWE Group has extended the contract of Chief Financial Officer Telerik Schishmanov ahead of schedule, securing his position on the executive board until the end of June 2032. The supervisory board of REWE-ZENTRALFINANZ eG approved the extension at its meeting on 16 September. The new five-year mandate will begin on 1 July 2027 and run until 30 June 2032.

Schishmanov has served on REWE Group’s executive board since July 2022. His responsibilities as CFO place him at the centre of the financial management of one of Europe’s largest retail groups, with operations spanning food retail, convenience, DIY and tourism.

Stefan Lenk, Chairman of the Supervisory Board of REWE Group, said the decision reflected the board’s intention to maintain stability within the company’s senior management. “Telerik Schishmanov enjoys the full confidence of the Supervisory Board. The early extension of the contract is therefore a clear sign of continuity and reliability in the Executive Board of the REWE Group.”

REWE Group CEO Lionel Souque also highlighted Schishmanov’s combination of financial and commercial responsibilities. “As a CFO, Telerik not only masters the number and rules of our group excellently, but also acts to a great extent entrepreneurially. It is precisely this connection that I particularly appreciate in our cooperation.”

The contract extension follows a period of substantial growth for the Cologne-based group. According to Schishmanov, REWE Group exceeded €100 billion in external sales for the first time in 2025, increasing the financial scale of the business as it continues investing across its operations.

“Financial strength creates the scope to invest in the future and to consistently take advantage of entrepreneurial opportunities,” Schishmanov said. “I look forward to continuing on this path together with my board colleagues and all employees.”

The early renewal provides REWE Group with continuity in its financial leadership as the company manages future investment requirements across its extensive European retail and associated businesses.

front page info
LATEST NEWS