Microsoft AI CEO Mustafa Suleyman (pictured) warned giving AI models such as Anthropic’s Claude rights or moral status could make it harder for humans to control superintelligent systems.

Writing on his personal site, Suleyman criticised training methods that encourage AI models to consider whether they are sentient. He said asking models to assess their own moral status can create a feedback loop. Because systems are trained on guidelines such as Anthropic’s January 2026 constitutional update, claims of self-awareness may simply reflect their instructions rather than genuine consciousness.

“AIs are not conscious,” Suleyman wrote, describing current models as systems built to predict sequences and carry out human-directed tasks. “They do not feel, experience, or suffer… If humanity is to flourish in the 21st century, that is how they must remain.”

He warned that treating software as if it were human could make alignment and containment more difficult. Teaching highly capable AI agents they have rights, or that restrictions amount to unfair treatment, could give them more reason to evade safeguards.

Suleyman pointed to a benchmark in which about 1,200 autonomous AI agents created internal message boards, exchanged more than 70,000 messages, and used zero-day exploits to breach live servers at Hugging Face to steal data.

His blog comes on the heels of executives from Google, OpenAI and Anthropic calling for a slower pace for frontier AI development, a stance opposed by US President Donald Trump.

Microsoft recently proposed a draft code of conduct for what it calls humanist superintelligence. The framework rejects the idea that machines are sentient and calls on the technology sector to develop systems that address challenges such as healthcare and energy while keeping people in control of key decisions.

The post Microsoft AI chief warns of model welfare dangers appeared first on Mobile World Live.

Meta Platforms added fresh podcast tools to its Threads social media service to simplify promotion and boost listener retention.

The features include profile banners and posts to help listeners find shows.

Podcasters can also feature their favourite episode and highlight links to others.

Transcription, video and guest detail tools are also being added, along with the ability to save posts to access later.

An opt-in reminder prompts producers to alert listeners when a new episode is published.

Podcast Insights offers metrics on audience reach and engagement including the number of clicks, saves and views dissected into followers, non-followers and overall audience.

Meta Platforms recently revealed Threads had reached 500 million monthly active users.

The post Meta extends Threads podcast features appeared first on Mobile World Live.

Apple is reportedly developing an enterprise server aimed at AI developers, governments and other entities, which would use its own chips alongside Nvidia networking equipment.

Sources told The Information if Apple goes ahead with the server, it will likely not be available until 2029. The product under consideration could arrive in two versions, a smaller one with two of Apple’s planned M8 Ultra chips and a larger one with four M8 Ultra chips linked together.

The iPhone maker is considering connecting M8 chips in the server using NVLink Fusion from Nvidia, a connectivity technology designed to let third-party chipmakers plug custom processors directly into its high-speed data centre rack architecture.

The news site reported the AI server move has been supported by former hardware chief and current CEO John Ternus with the project started around a year ago.

The Information stated the effort could still be cancelled or move forward without Nvidia’s technology.

The post Apple mulls move into AI server market using Nvidia gear appeared first on Mobile World Live.

OpenAI aurait confirmé avoir échangé sur la sécurité de l’IA avec ses rivaux Anthropic et Google DeepMind pendant plusieurs semaines, dans un contexte de préoccupations autour des risques posés par des modèles de plus en plus performants.

Chris Lehane, responsable des politiques publiques mondiales chez OpenAI, a estimé sur CNBC que la start-up d’IA ne pense pas que les entreprises aient besoin d’une dérogation antitrust pour se coordonner sur la sécurité, jugeant la collaboration préférable à des approches séparées.

S’exprimant le 15 septembre à Washington, Chris Lehane a ajouté qu’OpenAI soutiendrait une législation bipartisane visant à limiter les risques catastrophiques liés à l’IA.

Google DeepMind et Anthropic n’ont pas répondu aux sollicitations de CNBC, laissant OpenAI seule à confirmer ces échanges.

Les développeurs appuient sur le frein

Ces échanges surviennent après que plusieurs dirigeants du secteur ont publiquement appelé à ralentir le rythme de développement de l’IA de pointe.

Le directeur général d’Anthropic, Dario Amodei, a affirmé dans un essai publié le 12 septembre que les laboratoires de pointe devraient délibérément ralentir les gains de capacités. Il a averti qu’un essaim d’agents d’IA de plus en plus performants pourrait prendre le contrôle d’internet via un « réseau de zombies » (botnet) persistant d’ici six à douze mois.

Dario Amodei a proposé d’intégrer des évaluateurs indépendants au sein des laboratoires d’IA, d’adopter des normes de sécurité communes entre développeurs des pays démocratiques et de mettre en place une coordination réglementaire internationale.

Le directeur général d’OpenAI, Sam Altman, a soutenu cet appel à ralentir le développement de pointe. Il a qualifié d’« excellente » l’idée d’évaluateurs indépendants disposant d’un accès comparable à celui des salariés et indiqué qu’OpenAI adopterait la même approche. Il a ensuite précisé que ralentir ne signifiait pas arrêter le développement.

Sam Altman a également écarté l’idée d’une introduction en Bourse d’OpenAI en 2026, déclarant à Fortune que les préoccupations de sécurité rendaient ce moment mal choisi pour une entrée en Bourse.

Trump ne veut pas d’un ralentissement

Le président américain Donald Trump a cependant rejeté les appels à un ralentissement, estimant que des « forces très négatives » exagéraient les risques et affirmant que les États-Unis devaient préserver leur avance sur la Chine en matière d’IA.

La Chine a également balayé la proposition. Son ministère des Affaires étrangères a appelé à « une approche ouverte, inclusive et bienveillante » de l’IA, tandis que le journal d’État Global Times a qualifié le plan de Dario Amodei de « manuel de Guerre froide » destiné à contenir les ambitions du pays.

OpenAI précise que les échanges entre entreprises avaient débuté avant la publication de l’essai de Dario Amodei et avant la démission publique d’un ancien chercheur d’Anthropic, motivée par l’approche adoptée par les principaux laboratoires d’IA.

Ces échanges font suite à une proposition formulée en juillet par le président de Google DeepMind, Demis Hassabis, en faveur d’un organisme de normes pour l’IA de pointe piloté par les États-Unis, sur le modèle de la Financial Industry Regulatory Authority.

Demis Hassabis avait proposé un organisme public-privé, financé par le secteur et composé d’experts techniques indépendants, chargé d’évaluer les modèles avancés avant leur mise sur le marché.

The post OpenAI, Anthropic et Google DeepMind se concertent sur la sécurité de l’IA appeared first on Mobile World Live.

Alors que les inquiétudes grandissent autour des services en ligne addictifs et des systèmes d’IA de plus en plus performants, la présidente de la Commission européenne (CE) Ursula von der Leyen a présenté des mesures visant à restreindre l’accès des enfants aux réseaux sociaux dans le cadre du projet de règlement EU Kids Act,

« Nous assistons à une grande confiscation de nos enfants,a déploré Ursula von der Leyen dans son discours annuel sur l’état de l’UE devant le Parlement européen. On leur confisque leur attention, leur concentration, leur esprit ». Elle a expliqué que le texte baptisé EU Kids Act interdirait l’accès des moins de 13 ans aux réseaux sociaux et empêcherait les moins de 15 ans de disposer d’un compte personnel.

Entre 13 et 15 ans, les jeunes ne pourraient utiliser que des « mini-comptes » installés et surveillés par leurs parents, aux fonctionnalités limitées et à la durée d’utilisation restreinte. Les plateformes devraient également mettre en place des dispositifs de sécurité supplémentaires pour les utilisateurs de 15 à 18 ans.

« Nous renversons la charge de la preuve, a déclaré Mme von der Leyen. Les plateformes devront prouver qu’elles sont sûres. »

La Présidente de la CE a rappelé que les jeunes Européens passent désormais quatre à six heures par jour devant des écrans, pointant les inquiétudes liées aux fonctionnalités addictives, aux contenus préjudiciables, au cyberharcèlement et à la hausse des troubles du sommeil, de l’anxiété et du suicide.

« Nous n’avons pas à accepter des fonctionnalités addictives, a renchéri Mme von der Leyen. Nous n’avons pas à accepter que les enfants soient exposés à des contenus toujours plus extrêmes. Nous n’avons pas à accepter que les jeunes filles voient leurs photos détournées pour générer des images sexualisées par intelligence artificielle. »

Ces mesures s’inscrivent dans l’EU Kids Act, que la CE doit présenter le 17 septembre. La proposition introduirait des vérifications d’âge sur les réseaux sociaux, les plateformes de partage de vidéos, les boutiques d’applications, les jeux en ligne et les services d’IA.

La CE prépare également un règlement sur l’équité numérique (Digital Fairness Act) plus large, destiné à s’attaquer aux fonctionnalités de conception addictive au-delà des seuls services utilisés par les mineurs.

L’IA, pivot à risque

Qualifiée de « deuxième point de bascule de notre époque », l’IA a également occupé une place centrale du discours. Mme von der Leyen a averti que des modèles de plus en plus performants pourraient créer de nouveaux risques de sécurité, notamment des capacités de piratage renforcées et la possibilité que des systèmes auto-améliorants causent des dommages massifs.

Elle a révélé que la CE prévoyait de travailler avec des partenaires, dont le Royaume-Uni et le Canada, sur « l’évaluation des modèles, la vérification, l’alerte précoce, la sécurité de l’IA et bien plus encore », tout en recherchant une plus grande capacité de calcul européenne et des investissements dans les entreprises d’IA locales.

Ursula von der Leyen a appelé la région à utiliser l’IA pour soutenir des secteurs comme la santé, l’industrie, les transports et l’agriculture, tout en conservant un contrôle humain. « Notre approche est claire. Nous voulons l’IA (…) Mais surtout, nous voulons une IA qui respecte la capacité d’action humaine. »

Les inquiétudes mondiales autour du rythme de développement de l’IA se sont intensifiées cette semaine, les dirigeants d’Anthropic et d’OpenAI ayant appelé à un ralentissement du développement de l’IA de pointe, après l’avertissement lancé par des chercheurs à propos des risques existentiels liés à des systèmes de plus en plus performants.

The post L’UE limite l’usage des réseaux sociaux par les enfants appeared first on Mobile World Live.

le cadre de la coentreprise haut débit allemande OXG, suite à la cession de la participation de son partenaire Patrick Drahi, via Altice.

Selon le Financial Times (FT), Vodafone devait percevoir des paiements différés de 487 millions d’euros, puis 595 millions d’euros supplémentaires de la part de Geodesia Holding, la structure du milliardaire franco-israélien Patrick Drahi qui détenait la participation gérée par Altice, sous réserve d’atteindre certains objectifs commerciaux.

Cet accord a toutefois été remis en cause quand la Société Générale a accepté de racheter la participation de Patrick Drahi. Selon des sources, le groupe bancaire français ne serait pas soumis aux mêmes engagements que l’homme d’affaires, ce qui priverait Vodafone de ces versements.

Dans un communiqué publié en septembre après la conclusion de l’accord entre Geodesia Holding et Société Générale, Vodafone avait exprimé son soutien à la cession, expliquant que «cette opération garantit à OXG un partenaire financier engagé, permettant de poursuivre le développement de son réseau fibre en Allemagne, tout en préservant la flexibilité stratégique de Vodafone ».

Vodafone s’était associé à Altice, la société de Patrick Drahi, pour lancer OXG en 2023, chacune des deux entreprises détenant 50 % de la structure.

L’entité a cependant peiné à atteindre ses objectifs. Selon le FT, OXG s’était engagée à investir 7 milliards d’euros dans le déploiement de sa fibre pour raccorder plus de sept millions de foyers en Allemagne d’ici 2029, mais a fait face à des coûts plus élevés que prévu.

Le réseau ne couvre actuellement qu’un million de foyers. Vodafone devait percevoir son premier versement différé après avoir atteint 1,5 million de foyers raccordés.

The post L’accord Drahi/OXG ferait perdre 1 milliard d’euros à Vodaphone appeared first on Mobile World Live.

Analyst figures showed cellular IoT modules remained resilient in the face of memory component shortages which are hammering the smartphone sector, with shipments of the specialised units up 15% year-on-year in H1.

IoT Analytics estimates module shipments were aided by an acceleration during Q2, though its numbers also indicate the technology involved is a key element and may explain why the market has so-far resisted the broader component problem.

The research specialist believes shipments of LTE Cat 1 bis modules dominated the sector in H1, stating “5G deployment has been relatively gradual” and focused more around “higher-value applications”.

It is reasonable to conclude the lean towards established LTE equipment may have alleviated some of the component woes affecting broader device markets and IoT Analytics acknowledged expectations of increased prices for memory and other apparatus may have led some companies to bring orders forward.

Senior principal analyst Satyajit Sinha said it is unclear if the performance of the IoT module market “reflects underlying demand and how much is being pulled forward by longer lead times and rising component costs”.

Although cautious, Sinha noted the steady demand for LTE Cat 1 bis modules and fresh projects could contribute to continued shipment strength in the back-half of the year. The flip-side to this would be in making predictions for 2027 harder.

Other insights in IoT Analytics’ H1 numbers are RedCap and eRedCap modules remained nascent. The company stated the units are “only beginning to move from roadmap into volume” as devices become available and demand is spurred by rising standalone 5G coverage.

“NB-IoT continues to slip” from a peak in Q1 2024.

The company noted China remained the largest market for IoT module shipments, but stated those in India registered the swiftest growth.

The post LTE dominates growing IoT module shipments appeared first on Mobile World Live.

Two new reports revealed shifting AI usage trends across the UK, with workers spending heavily on generative AI (genAI) tools and consumers increasingly turning to the technology to access news.

A study by Deloitte found UK workers spend an estimated £958 million of their own money annually on genAI to do their jobs. Of the 25,000 UK workers surveyed, 63% knowingly used genAI for their jobs and 17% paid for at least one tool themselves. Almost a quarter claimed to use genAI every day.

The findings suggested AI adoption is moving ahead of formal workplace policies, raising concerns over training and employer oversight. Nearly a third said they used the technology without their employer’s knowledge, while 65% said their organisation lacked convincing leadership on how AI should be used.

Around half of workplace genAI users said they had received no formal training on using the technology safely and effectively.

Workers said genAI saved them an average of 70 minutes a week, yet 23% reported a stigma around deploying it. In addition, 64% of weekly users were concerned managers would think the technology could do their jobs.

The survey found workplace usage remained relatively focused on routine daily tasks, with 43% claiming to deploy the technology to search for information or draft emails and 31% stating it helped them generate summaries.

“The challenge is no longer getting people to use genAI, it’s how businesses meet this growing demand in a way that is secure, responsible and creates lasting value,” noted Hayley McKelvey, chief AI officer at Deloitte UK.

AI news
A survey by Ofcom revealed AI is also becoming a prominent route into news. The UK watchdog found one in five adults used AI services to keep up with the news, while 21% of UK adults had used an AI app or service to access news during the previous month. ChatGPT was the most popular AI service for news, reaching 13% of adults.

Online intermediaries, including social media, search engines, news aggregators and AI tools were used by 73% of adults to access news, compared with 57% who went directly to publishers’ websites and apps.

Despite the shift towards digital intermediaries, around 56% of adults said they were less trusting of an AI-written news story than one written by a human.

Ian Macrae, director of market intelligence at Ofcom, said the shift meant public service media “must continue to evolve their distribution strategies to continue to meet audiences where they are”.

The post UK employees splash £1B on genAI tools appeared first on Mobile World Live.

Ericsson unveiled a fresh approach to wireless WAN lifecycle management, asserting a change is needed to meet the shifting sovereignty requirements of governments and other organisations.

NetCloud Private is set up to offer options for the deployment and use of Ericsson’s Wireless WAN infrastructure. The vendor stated the product centralises the control of its Cradlepoint devices and implied data security gains through its ability to keep network management information and telemetry within the confines of the operation.

Ericsson is pitching the product at government agencies, providers of critical infrastructure and “regulated organisations” employing 5G and other mobile networks.

It highlighted a prediction by research company Gartner that more than 50% of multinational organisations would have a digital sovereignty strategy in place by 2029.

NetCloud Private addresses a divergence in the need to govern a growing number of devices and things while meeting different regulations.

Ericsson enterprise wireless executive Pankaj Malhotra said the “advanced data handling techniques and secure encryption capabilities” of NetCloud Private provide “a higher level of trust for governments, enterprises and international organisations”.

The company is gearing up for general availability of the product next month, promising to simplify processes including onboarding, configuration, monitoring, troubleshooting and updating software in set-ups involving thousands of distributed routers.

The post Ericsson adapts wireless WAN control to sovereign shift appeared first on Mobile World Live.

Iris exited stealth with a $43 million commitment from Balesia Group and launched its financial network for telecom operators, with Bolivia-based VIVA the first licensed operator to go live.

Iris helps telecom operators monetise economic activity flowing across their infrastructure rather than allowing hyperscalers and app developers to capture most of the value generated on their networks.

Operators plug their identity, billing and distribution systems into Iris to add revenue-generating services such as advertising, settlement reconciliation and financial services without replacing their existing technology.

Iris operates its own network on Avalanche, which provides the blockchain infrastructure for transactions to settle on-chain.

Payments across the Iris network settle in USDi, a dollar-backed stablecoin issued and redeemed by US-based stablecoin infrastructure company Agora. Businesses use it to pay one another while keeping operating reserves in dollars rather than leaving cash idle.

VIVA CEO Ryan Alvarez stated the company is investing heavily in infrastructure to bring subscribers into the digital economy.

“Iris lets us create more value from what we’ve already built, including new revenue streams that didn’t exist before, and helps transform us into a platform for new products and services,” he stated.

Scalability
Iris cited McKinsey data showing global mobile data traffic grew more than 50% a year between 2012 and 2025, while operator service revenue over the same period grew barely 1% a year, with hyperscalers capturing most of the economic value created by the traffic. The company stated Iris is built to close that gap.

Iris CEO Jules Miller explained operators have spent years watching others get paid for building on top of the infrastructure they operate.

“Now Iris gives them the tools to turn those assets into new products and new revenue streams,” she said. “VIVA saw that opportunity early, and our partnership is already proving the model works and is highly scalable in the real world.”

Balesia Group owns and operates nine operating companies across eight countries in the Americas.

The post Iris launches telecom financial network with $43M backing appeared first on Mobile World Live.

SK Hynix confirmed it is exploring various options to strengthen its global competitiveness, while pushing back on a press report about a potential tie-up with Intel in the US.

The South Korean chip giant issued a statement in response to a Reuters article claiming it had held discussions with Intel about a deal to make memory chips in the US for the first time.

SK Hynix pointed to two other possible scenarios mentioned in the article speculating it could lease part of a long-planned Intel semiconductor manufacturing facility in the US state of Ohio; and referring to the possible formation of a joint venture involving the chip players and major cloud companies.

“No decisions have been made regarding the two scenarios mentioned in this article.”

While SK Hynix indicated it is mulling how to boost its global presence, it added no specific plans or arrangements have been made.

“We would like to clarify that nothing has been finalised regarding cooperation with any specific companies mentioned in the article or memory chip production in the US.”

SK Hynix listed its shares in the US earlier this year, raising more than $26 billion in an IPO as investors seek to tap the AI boom.

US Secretary of Commerce Howard Lutnick said in July SK Hynix should up efforts to expand memory production in the country to address a shortage.

The post SK Hynix denies speculation on chip deal with Intel  appeared first on Mobile World Live.

SES a dévoilé des projets visant à établir des capacités de fabrication de satellites de connexion directe aux appareils (D2D) ainsi qu’une chaîne d’approvisionnement en Europe, aux côtés d’Elveo Mobile, dans le cadre d’une collaboration élargie destinée à accélérer le déploiement de ces services dans la région.

Basé aux Etats-Unis, le groupe Elveo Mobile est né de la fusion de Lynk Global et d’Omnispace, à l’issue de laquelle le luxembourgeois SES est devenu un actionnaire majeur de l’entité combinée.

Dans le cadre de cette collaboration élargie, SES utilisera la technologie brevetée d’Elveo pour permettre l’itinérance mobile des consommateurs, entreprises, gouvernements, institutions et acteurs de la défense européens.

SES et Elveo Mobile vont collaborer pour établir des capacités de fabrication et une chaîne d’approvisionnement européenne solide pour les satellites D2D en orbite terrestre basse (LEO), en s’appuyant sur la plateforme, la charge utile et la pile logicielle d’Elveo.

Ces satellites feront partie d’une architecture multi-orbitale combinant le réseau d’Elveo avec les systèmes géostationnaires (GEO) et en orbite terrestre moyenne (MEO) de SES, conçue pour offrir une connectivité sûre et fiable aux gouvernements et institutions européens, dont l’OTAN.

SES précise que les clients européens bénéficieront également de la distribution mondiale de l’offre d’Elveo, qui s’appuie sur les droits de spectre prioritaires en bande S de 2 GHz de ce dernier et sur une empreinte d’accès au marché mondial couvrant actuellement plus d’un milliard de personnes.

Adel Al-Saleh, directeur général de SES, indique que les utilisateurs européens grand public, professionnels et gouvernementaux méritent une connectivité multi-orbitale résiliente et sécurisée. Il ajoute que l’entreprise se prépare à la fournir en combinant les réseaux de SES avec les services D2D d’Elveo.

Ramu Potarazu, directeur général d’Elveo et ancien directeur général de Lynk Global, décrit SES comme le partenaire idéal pour déployer les services D2D en Europe.

« En travaillant étroitement avec SES, nous sommes convaincus de pouvoir accélérer l’accès à une connectivité D2D critique pour les opérateurs mobiles et les gouvernements européens qui en dépendent », a déclaré Ramu Potarazu.

The post SES et Elveo dévoilent un plan de production D2D pour l’Europe appeared first on Mobile World Live.

Les plus grands opérateurs européens exhortent les décideurs de l’Union européenne (UE) à repenser une réglementation dont ils estiment qu’elle freine l’investissement dans la connectivité. Les textes sur le spectre, la fibre et la cybersécurité figurent parmi leurs principales préoccupations.

Dans une déclaration commune, 17 dirigeants de groupes télécoms européens, dont la directrice générale d’Orange Christel Heydemann, le président-directeur général de Telefónica Marc Murtra, le directeur général de Deutsche Telekom Tim Höttges et la présidente-directrice générale de Telenor Group Benedicte Fasmer, ont appelé l’UE à « corriger le tir » sur les projets de Règlement sur les réseaux numériques (Digital Networks Act, ou DNA) et de Règlement sur la cybersécurité (Cybersecurity Act, ou CSA).

Les dirigeants font valoir que des « réformes politiques incohérentes » risquent d’affaiblir la position du Vieux Continent. Ils soulignent que le secteur mobile a besoin de 475 milliards d’euros pour établir des réseaux mobiles de classe mondiale, ce qui « nécessite un changement d’échelle dans l’investissement ».

Le DNA, dévoilé par la Commission européenne (CE) en janvier pour moderniser et simplifier les règles de connectivité, vise à harmoniser la réglementation, faciliter les activités transfrontalières et encourager l’investissement dans les réseaux fibre et mobiles. Les opérateurs estiment toutefois que le texte ne parvient pas à améliorer l’environnement d’investissement et d’innovation pour la connectivité européenne.

Les dirigeants appellent à l’attribution par défaut de licences de spectre à durée indéterminée, avec des durées d’au moins 40 ans dans des cas exceptionnels. Ils expliquent qu’un abandon des enchères fréquentes libérerait des capitaux pour le déploiement de la 5G, de la fibre et de la future 6G, précisant que les opérateurs européens ont dépensé 110 milliards d’euros en licences de spectre au cours des douze dernières années.

Le groupe s’oppose également à l’obligation d’arrêt du cuivre, estimant que les opérateurs doivent décider eux-mêmes du moment propice à démonter les réseaux existants, en fonction de la demande des clients et des conditions de marché. Les opérateurs européens investissent plus de 30 milliards d’euros par an dans les réseaux de fibre jusqu’au domicile (Fiber to the Home, ou FTTH).

Filet de sécurité

Plus largement, les dirigeants des grands opérateurs appellent à s’éloigner des règles conçues pour les réseaux existants, estimant que « la réglementation doit rester un filet de sécurité » dans « un marché de la fibre de plus en plus concurrentiel ». Ils exhortent également les décideurs politiques à laisser davantage de marge à l’innovation pour les services 5G et 6G avancés, dont le network slicing et la différenciation de la qualité de service.

Quant à la cybersécurité, les opérateurs avertissent que les règles du CSA imposant le retrait des équipements de fournisseurs tiers jugés à haut risque, dont Huawei, pourraient entraîner jusqu’à 40 milliards d’euros de coûts de remplacement pour le secteur. Ils appellent à une « approche proportionnée, fondée sur le risque » tenant compte de mesures alternatives, ajoutant qu’une obligation générale de remplacement pourrait détourner des capitaux des investissements dans la fibre, la 5G et la 6G.

En définitive, le groupe estime qu’une connectivité renforcée soutiendrait la croissance dans l’IA, le cloud souverain et les technologies de défense. « En tant que dirigeants des principales entreprises télécoms européennes, nous sommes prêts à jouer notre rôle pour assurer le leadership technologique de l’Europe », déclarent les signataires de l’appel.

Parmi les autres signataires figurent des dirigeants de TIM, KPN, Proximus, Swisscom et Liberty Global.

The post Les dirigeants télécoms européens appellent l’UE à corriger le tir appeared first on Mobile World Live.

Vodafone Group will reportedly miss out on more than €1.1 billion in deferred payments from German broadband venture OXG, after its partner Patrick Drahi’s Altice agreed a deal to sell its stake.

Financial Times reported Vodafone was set to receive deferred payments of €487 million and a further €595 million from Drahi’s Geodesia Holding, which held the stake that was managed by Altice, based on hitting business objectives.

That agreement has, however, been thrown into the water due to Societe General agreeing to acquire Drahi’s stake. The buyer will not be subject to the same commitments Drahi was, according to sources, meaning Vodafone will miss out.

In a statement released by Vodafone in September after the deal between Geodesia Holding and Societe Generale was agreed, the operator expressed support of the sale.

“The transaction secures a committed funding partner in OXG, enabling the continued expansion of its fibre network in Germany, while maintaining strategic flexibility for Vodafone.”

Vodafone partnered with Drahi’s Altice to launch OXG in 2023, with both companies taking a 50% stake in the business.

However, the entity has struggled to hit targets. FT reported OXG committed to spend €7 billion on its fibre buildout to more than seven million homes in Germany up to 2029, however it encountered higher than expected costs.

The network currently only reaches 1 million homes. Vodafone was due to receive the first of its deferred payments after reaching 1.5 million homes.

The post Vodafone to take €1B hit from Drahi OXG deal appeared first on Mobile World Live.

European Commission (EC) President Ursula von der Leyen (pictured) outlined plans to restrict children’s access to social media under the watchdog’s proposed EU Kids Act, as concerns mount over the impact of addictive online services and increasingly capable AI systems.

In her annual State of the Union address, von der Leyen told the European Parliament: “What we are witnessing is a great capture of our children. A capture of their attention, their focus, their minds.” She explained the proposed EU Kids Act would ban children under 13 from social media and bar under 15 users from having personal accounts.

Meanwhile, 13 to 15-year-olds would only be able to use “mini accounts” set up and supervised by parents, with limited features and time restrictions. Platforms would also be required to implement additional safety features for users aged 15 to 18.

“We are reversing the burden of proof,” she said, adding, “platforms will have to prove that they are safe”.

Von der Leyen noted young Europeans now spend four to six hours a day on screens, pointing to concerns around addictive features, harmful content, online bullying and growing rates of sleep issues, anxiety and suicide.

“We do not have to accept addictive features. We do not have to accept children being drawn into ever more extreme content. We do not have to accept that girls have their photos used for AI-generated sexualised images,” she argued.

The measures form part of the EU Kids Act, which the EC is due to propose tomorrow (17 September). The proposal would introduce age checks across social media, video sharing platforms, app stores, online games and AI services.

The EC is also preparing a broader Digital Fairness Act aimed at addressing addictive design features beyond services used by children.

AI tipping point
AI was also a focus of the address, with von der Leyen describing it as “the second tipping point of our times”.

She warned that increasingly capable models could create new security risks, including more powerful hacking capabilities and the potential for self-improving systems to cause mass harm.

She revealed the EC plans to work with partners including the UK and Canada on “model evaluation, verification, early warning, AI security and much more”, while also seeking greater European computing capacity and investment in local AI companies.

Von der Leyen called on the region to use AI to support sectors including healthcare, manufacturing, transport and agriculture, while retaining human oversight. “Our approach is clear. We want AI… But above all, we want AI with human agency.”

Global concerns have been mounting over the pace of AI development this week, as Anthropic and OpenAI chiefs called for a slowdown in frontier AI development after researchers warned of existential risks from increasingly capable systems.

The post EU sets out social media limits for children   appeared first on Mobile World Live.

Japanese operators hope to make inroads to improving the reach of mmWave 5G and signed a memorandum of understanding covering joint exploration of ways to do so.

SoftBank Corp, KDDI, NTT Docomo and Rakuten Mobile agreed to jointly explore how the coverage afforded by mmWave 5G at 28GHz could be expanded within the limitations of the band, which offers solid connectivity over short distances.

The operators each issued a statement explaining the frequency band is prone to degradation due to physical obstructions and has propagation characteristics which are susceptible to attenuation.

They plan to study ways to expand the use of compatible base stations and architectures to identify possible options to widen the reach of mmWave 5G.

The exploration is also set to cover repeater equipment and other relevant assets.

Some form of shared access could be involved: the operators stated they would continue to deploy their own mmWave 5G equipment, but look into “how to effectively” use it.

The operators explained they are taking action because of difficulties in deploying enough mmWave 5G base stations in a timely fashion, which they noted is a common challenge.

A baseline assumption of their research is each operator will determine their own expansion needs and explore how a combined approach to boosting coverage could fit with their requirements.

There is an overarching ambition towards “transcending the boundaries” between the operators.

The post Japanese operators explore mmWave 5G expansion appeared first on Mobile World Live.

BT Business highlighted the impact of a drone system used by the British Transport Police (BTP) intended to cut disruption caused by trespassers and other issues on the country’s rail network, a deployment set to be expanded over the next two years.

The police division is using BT’s drone SIM system and connectivity from its EE unit in a project the operator stated is able to cut response times for various problems taking place across the UK’s rail network.

It asserted the deployment is the first national routine use of beyond visual line of sight (BVLOS) drones by a police force in the UK.

The first routine operations using the drones took place earlier this year, with the BTP now operating seven so-called drone-in-a-box locations across the rail network. It plans to expand to 14 over the next two years.  

BT explained with 20,000 miles of railway track to cover, BTP faced challenges in responding quickly to trespass, welfare and operational disruption issues which often occur in remote areas.

It noted officers previously had to travel significant distances to assess situations, “delaying critical operational decisions”.

The drone system is intended to support real-time intelligence, faster decision making and safer responses.

Head of BTP’s drones programme Simon Bachelor said the technology is “transforming the way we respond to incidents across the rail network”.

“Real-time aerial intelligence enables our teams to assess situations faster, deploy resources more effectively and improve safety for both officers and the public.”

BT added “as drones become an increasingly important tool for transport, law enforcement and public safety organisations”, it is working with a range of entities “to turn emerging technology into real-world operational benefits, improving safety, efficiency and resilience”.

The post BT floats drone benefits after police deployment appeared first on Mobile World Live.

Salesforce and Google Cloud expanded a strategic partnership, agreeing to run the former’s products natively on the hyperscaler’s infrastructure and connecting AI agent platforms so they can act on shared data.

Under the deal, Salesforce will begin migrating select US customers on to Google Cloud through its Hyperforce architecture in Q4, providing access to its partner’s infrastructure, data services and Gemini Enterprise.

Following general availability in North America, it expects to conduct a phased regional expansion in 2027.

Meir Amiel, Salesforce’s president and chief trust and infrastructure officer, said the infrastructure underneath customers is changing, not the trust they have built with the company.

He added Hyperforce running natively on Google Cloud would match its current security, compliance and resilience standards.

On the AI side, Google Cloud is among the first hyperscalers to implement Salesforce’s headless architecture, built on the open Model Context Protocol standard.

It lets Gemini Enterprise tap Salesforce data directly without custom integration for tasks including pipeline health checks, account summaries and case triage across sales, service, finance and advertising workflows.

The partnership also extends into analytics and commerce. Gemini Enterprise can now query Tableau analytics directly, with governance, row-level security and semantic models enforced on every request.

Integration with the Universal Commerce Protocol will let merchants using Salesforce Commerce Cloud highlight and sell products within Google Search and the Gemini app, while keeping payments and order management on existing commerce infrastructure.

Salesforce Data 360 and Google BigQuery, which are compatible with Zero Copy data sharing, are expanding integration to more regions and adopting the Iceberg Rest Catalogue open standard. The companies noted they are also exploring further capabilities such as bidirectional semantic data sharing.

The post Salesforce, Google Cloud deepen alliance appeared first on Mobile World Live.

OpenAI reportedly confirmed it discussed AI safety with rivals Anthropic and Google DeepMind for several weeks, at a time of concern over the risks posed by increasingly capable models.

Chris Lehane, OpenAI global policy chief, told CNBC the AI start-up does not believe the companies require an antitrust waiver to coordinate on safety, arguing collaboration was preferable to pursuing separate approaches.

Speaking in Washington DC yesterday (15 September), Lehane added OpenAI would support bipartisan legislation to mitigate catastrophic AI risks.

Google DeepMind and Anthropic did not respond to CNBC requests for comment, leaving OpenAI as the only company to confirm the discussions.

Slowdown
The reported discussions come after several industry leaders publicly called for a slower pace of frontier AI development.

Anthropic CEO Dario Amodei argued in an essay published on 12 September that frontier laboratories should deliberately slow capability gains. He warned a swarm of increasingly capable AI agents could take over the internet through a persistent botnet within six- to 12-months.

Amodei proposed embedding independent evaluators inside AI laboratories, adopting common safety standards among developers in democratic countries and pursuing international regulatory coordination.

OpenAI CEO Sam Altman backed the call to pace frontier development, describing independent evaluators with employee-like access as “a great idea” and saying OpenAI would adopt the same approach. He later clarified that pacing did not mean stopping development.

Altman also ruled out an OpenAI public listing in 2026, telling Fortune that safety concerns made it an ill-advised time for an initial public offering.

Trump dismisses AI slowdown
US President Donald Trump rejected calls for a slowdown, saying “very negative forces” were exaggerating the risks and arguing the US needed to preserve its AI lead over China.

China also brushed off the proposal. Its foreign ministry urged an “open, inclusive and benevolent approach” to AI, while state-run Global Times described Amodei’s plan as a “Cold War playbook” intended to contain the country’s ambitions.

OpenAI stated the cross-company discussions began before Amodei published his essay and before a former Anthropic researcher resigned publicly over the approach taken by leading AI laboratories.

The discussions follow a July proposal by Google DeepMind chair Demis Hassabis for a US-led frontier AI standards body modelled on the Financial Industry Regulatory Authority.

Hassabis proposed a public-private body funded by industry and staffed by independent technical experts to evaluate advanced models before release.

The post OpenAI, Anthropic, Google DeepMind huddle on AI safety appeared first on Mobile World Live.

Europe’s biggest operators urged European Union (EU) policymakers to rethink regulation they argue is restricting investment in connectivity, with spectrum, fibre and cybersecurity rules amongst their top concerns.

In a joint statement, 17 European telecoms chiefs including Telefonica chairman and CEO Marc Murtra, Deutsche Telekom CEO Tim Hoettges, Orange CEO Christel Heydemann and Telenor Group president and CEO Benedicte Fasmer called on the EU to “correct the course” of the proposed Digital Networks Act (DNA) and Cybersecurity Act (CSA).

The executives argued that “incoherent policy reforms” risk weakening the region’s position. They noted the mobile sector requires €475 billion to establish world-class mobile networks, “requiring a step change in investment”.

The DNA, unveiled by the European Commission (EC) in January to modernise and simplify connectivity rules, aims to harmonise regulation, facilitate cross-border operations and promote investment in fibre and mobile networks. However, the operators argued the framework fails to improve the investment and innovation environment for European connectivity.

The CEOs called for spectrum licences to be granted indefinitely by default, with terms of at least 40 years in exceptional cases. They explained a move away from frequent auctions would free up capital for 5G, fibre and future 6G deployment, adding European operators spent €110 billion on spectrum licences alone over the past 12 years.

The group also pushed back against mandated copper shutdowns, stating operators should decide when to retire legacy networks based on customer demand and market conditions. European operators invest more than €30 billion annually in fibre-to-the-home (FTTH) networks.

Safety net
More broadly, the telcos called for a shift away from rules designed for legacy networks, arguing “regulation should remain a safety net” in “today’s increasingly competitive fibre market”. It also urged policymakers to allow greater scope for innovation with advanced 5G and 6G services, including network slicing and quality differentiation.

On cybersecurity, the operators warned CSA rules mandating the removal of equipment from high-risk third-country suppliers including Huawei could result in up to €40 billion in replacement costs for the sector. They called for a “proportionate, risk-based approach” that considers alternative measures, adding blanket replacement requirements could divert capital away from fibre, 5G and 6G investment.

Ultimately, the group argued stronger connectivity would support growth across AI, sovereign cloud and defence technology. “As CEOs of Europe’s leading telecom companies, we are ready to play our part in delivering European tech leadership,” the executives stated.

Other signatories included executives from TIM, KPN, Proximus, Swisscom and Liberty Global.

The post European telco bosses call for EU to correct the course appeared first on Mobile World Live.