Germany-based robotics firm Agile Robots reportedly plans to produce up to 2,000 humanoid robots next year through a fresh partnership with Nvidia, targeting universities and corporate research labs.

The collaboration was announced at Nvidia’s GTC conference in Berlin. Agile CEO Zhaopeng Chen told Reuters the pair are set to begin production of the Agile ONE S and ONE SN humanoid robot models by the end of 2026, expecting to manufacture between 1,000 and 2,000 units. Stanford University and ETH Zurich are among potential customers.

The Agile ONE SN, with the N standing for Nvidia to denote the chipmaker-specific configuration, combines Agile’s humanoid hardware with Nvidia’s Jetson Thor onboard computing platform and Isaac GR00T robotics AI model. It features five-fingered hands, cameras and other sensors to support dexterous manipulation and interaction with the physical environment.

Standing 1.4 metres tall and weighing around 40 kilograms, the robot is designed as a smaller alternative to full-sized humanoids. Nvidia said the robot intends to simplify the integration of hardware, simulation, training and deployment tools on a single platform, “cutting setup time from weeks to days”.

Spencer Huang, Nvidia’s product director for robotics, told Reuters the standardised configuration could save researchers two or three months of work procuring separately purchased components. “If we can bring researchers something that’s already 80% of the way there, it accelerates their development,” he said.

Nvidia said Agile’s ONE SN is scheduled to become available in early 2027, as competition intensifies among European humanoid developers.

ChatGPT moment
Alongside the partnership, the chipmaker also revealed its expanded robotics development platform, Nvidia Isaac, equipped with new tools for training and deploying AI-powered machines. The expanded platform includes Isaac OS, an open-source runtime framework, and Isaac Lab 3.0, a robot learning framework designed to train and refine robot skills. The company also introduced new reference workflows and simulation resources to speed up development.

“General-purpose robotics is advancing at extraordinary speed — the ChatGPT moment for robotics is fast approaching,” noted Jensen Huang, CEO of Nvidia, adding the platform would give developers “the models, data, simulation tools and computing systems to bring physical AI to the world’s largest industries”.

The chipmaker added European robotics players including UK-based Humanoid, French start-up Hugging Face and Swedish firm ABB Robotics are applying its technologies including Nvidia Isaac to industrial autonomous machines.

The post Agile Robots, Nvidia target up to 2,000 humanoids in 2027 appeared first on Mobile World Live.

Jio Platforms’ $3.2 billion IPO was reportedly fully subscribed within around 100 minutes of opening on Wednesday (21 October), as investors rushed to take part in India’s largest share sale.

Sources told Reuters institutional investors covered their allocated portion within 10 minutes of bidding opening, although demand from retail investors was slower. Around a third of the shares are reserved for retail buyers.

“It’s the fastest in the history of India,” a source said.

Reliance Industries’ telecoms unit is offering 270 million shares priced at between INR1,065 ($11.28) and INR1,119 apiece, with the offer closing on Friday (23 October). The IPO values the business at up to $108.3 billion and marks the first public offering from billionaire Mukesh Ambani’s Reliance conglomerate in two decades.

The company plans to use most of the IPO proceeds to repay or prepay borrowings at Reliance Jio Infocomm, its telecoms subsidiary, with the remainder earmarked for general corporate purposes, including capital expenditure.

Jio Platforms reported a profit of around $3.2 billion on revenue of $15.5 billion in the year to March.

The offering also comes despite a weak domestic equity market, after the country’s Nifty 50 benchmark index of 50 leading Indian companies slumped this year.

However, Indian wealth management firm SMIFS Limited noted “the long-term outlook for Jio Platforms Limited remains highly favourable, driven by sustained growth in India’s digital economy”.

According to the company, Jio Platforms will “benefit from rising 5G adoption, increasing smartphone penetration, accelerating broadband penetration, growing enterprise digitalisation and expanding demand for cloud and AI-enabled services”.

Jio’s telecoms business has more than 500 million subscribers, making it the world’s second largest telecoms operator in a single country after China Mobile. It disrupted India’s telecoms market after launching in 2016 with low-cost data plans, putting pressure on rivals Bharti Airtel and Vodafone Idea.

However, the company continues to face regulatory uncertainty and growing competition in satellite connectivity, as billionaire Elon Musk’s Starlink prepares to enter India’s market.

The post Jio Platforms launches long-awaited India listing appeared first on Mobile World Live.

Veon agreed a deal to acquire Uzbekistan mobile operator Perfectum for around $207 million, in a move designed to boost its spectrum holdings and accelerate 5G deployment in the country.

The group stated it had reached an agreement to buy 100% of Rubicon Wireless, which operates under the Perfectum brand. The deal includes around $93 million in existing Perfectum debt. Completion is expected by the end of the month, with regulatory approvals already secured.

The acquisition brings fresh spectrum and network infrastructure into Veon’s regional telecoms business, Beeline Uzbekistan. Perfectum holds 115MHz of spectrum across the 850MHz and 3,800MHz bands, alongside network equipment and a standalone 5G network deployed using Nokia technology.

Veon said the acquisition would increase Beeline Uzbekistan’s spectrum holdings by around 50%, giving the operator additional resources to expand network capacity and support future 5G and fixed wireless access services.

The companies also anticipate expenditure savings and improved “efficiency of future network investment” from integrating Perfectum’s network with Beeline Uzbekistan’s existing infrastructure.

The deal adds to Veon’s investments in Uzbekistan, a market of nearly 40 million people where demand for connectivity and digital services is apparently booming. The group said the combined assets would enable broader access to broadband to support mounting demand for financial services, digital applications and sovereign AI across the country.

“Uzbekistan is one of Veon’s core growth markets, and this acquisition reflects our confidence in its digital future,” noted Veon group CEO Kaan Terzioglu, adding the company aims to turn the deal into “stronger customer experiences, more efficient network investment and sustainable growth”.

The post Veon beefs Uzbek presence with $200M deal appeared first on Mobile World Live.

Elon Musk (pictured) revealed plans to give a SpaceX share to every child in the US next year, making the pledge a day after AI company Anthropic outlined plans to dedicate part of its equity to public benefit initiatives.

“Next year, I will give one of my SpaceX shares to every kid in America,” Musk posted on social media platform X, adding: “This will be the first of many.”

He provided no details on how the shares would be distributed or how the plan would be funded. According to Census Bureau data, around 73 million children live in the US. This would put the potential value of the giveaway at approximately $10.7 billion based on SpaceX’s closing share price of $146.03 on 16 October.

A regulatory filing made in August showed the billionaire held a 48.4% stake in SpaceX as of 30 June. The company went public earlier this year, raising $75 billion in a record-breaking IPO which briefly made the tech mogul the world’s first trillionaire.

Public Benefit Fund
Separately, Anthropic announced on Sunday (18 October) that it would commit 5% of its shares to a new Public Benefit Fund, with the potential for the allocation to grow as the company’s valuation increases.

The AI player stated the fund could receive an additional 2% of its shares for every 50% increase in its valuation above a baseline yet to be disclosed. Additional allocations would be made at the board’s discretion.

Anthropic stated the fund would give it “increased flexibility and scale” to pursue its mission, including “large donations to traditional non-profits or direct distributions to specific populations”.

One potential focus is AI-driven disruption to employment. The company stated that, although the technology could generate “enormous economic value”, it could “cause instability for many workers”. It has already committed $200 million to research and initiatives addressing the issue, “but given the pace of AI progress, bolder approaches may be needed”.

“The Fund will have the versatility to pursue them.”

Anthropic, which has confidentially filed for an IPO, stated further details on the fund’s governance and valuation thresholds would follow in the coming months.

The post Musk pledges SpaceX shares to every US child appeared first on Mobile World Live.

Microsoft reportedly withdrew from a deal to lease computing capacity at AI infrastructure player Nscale’s planned data centre in the UK due to delays related to power supply, adding to mounting concerns around the status of the country’s AI projects.

According to The Telegraph, the US technology giant pulled the plug on the mooted data centre to be constructed in Loughton, Essex, which was originally expected to receive a grid connection by the end of 2026.

However, it now faces a wait until the following decade for supply.

“Our relationship with Microsoft remains strong and long-term, and this unfortunate London-wide grid issue has no impact on our other contracts,” an Nscale spokesperson told Reuters. The pair are still collaborating across other projects, including developments in Norway, Portugal and the US.

The company reportedly added it remained “committed to both the Loughton project and the UK” and was exploring alternative power sources.

Announced last year, the Loughton development was backed by a £2.5 billion investment commitment from Nscale and formed part of a wider UK push to attract AI infrastructure spending. Microsoft had planned to lease the site’s entire 50MW capacity and deploy an AI supercomputer equipped with around 23,000 Nvidia GB300 chips.

The setback revives questions about the delivery of the UK’s headline AI investment pledges. A Guardian investigation in March raised concerns over the status of major announced projects, including the Loughton site, which was still being used as a scaffolding yard when visited by reporters in February. Nscale subsequently said it had purchased the site in September 2025 and was targeting delivery in 2027.

Power availability remains a wider constraint for UK data centre development, as growing demand from AI workloads puts pressure on grid infrastructure.

Nscale, which is backed by Nvidia, filed for a US IPO last month and raised $3.4 billion in pre-IPO financing. The UK-based company named Microsoft and Anthropic as expected major customers in its prospectus.

The post Microsoft pulls out of Nscale UK data centre deal appeared first on Mobile World Live.

Apple’s first foldable smartphone is facing a supply crunch within minutes of preorders opening, with delivery estimates for the iPhone Duo slipping into January 2027.

Preorders opened at 8am in New York on 16 October, ahead of the device’s scheduled retail launch on 23 October. By then, Apple’s online store was showing delivery dates that crossed into late January across both colour options and all storage tiers, from 256GB to 2TB.

Customers across social media platforms including Reddit and X reported delays despite following Apple’s advice to reserve their preferred colour and storage configuration in advance. According to Bloomberg, retail stores also exhausted their initial preorder stocks almost immediately.

It remains unclear whether the delays reflect unanticipatedly strong demand, limited supply or a combination of both. Apple typically reserves some units for walk-in customers on launch day, potentially offering buyers a way to secure the device sooner.

Unveiled last month as part of its first major product launch under new CEO John Ternus, the iPhone Duo marks Apple’s long-awaited entry into the foldable smartphone segment, which has so far remained a relatively niche category despite offerings from established rivals. The device starts at $1,999 for the 256GB model, making it the vendor’s most expensive iPhone to date.

Last month, Counterpoint Research estimated Apple could sell around 6 million iPhone Duo units in 2026, potentially providing fresh momentum for the foldable market. At the time, Counterpoint Research senior analyst Ivan Lam noted the number was “contingent on how well the company ramps up production of the new hardware”.

The post Apple iPhone Duo delivery dates slip into 2027 appeared first on Mobile World Live.

Mobile World Live (MWL) brings you our top three picks of the week as Apple ramped up its smart home ambitions, AT&T agreed a $2 billion sale of its Mexican business and Nokia offloaded its Enterprise Campus Edge unit.

Apple revamps smart home strategy with HomeView

What happened: Apple unveiled a new smart home hub and a range of connected devices as it stepped up efforts to challenge Amazon and Google in the market.

Why it matters: The centrepiece of its launch was the HomeView Hub, a smart display featuring the company’s upgraded Siri AI. Apple also partnered with LG Electronics and Schneider Electric to introduce the Apple Home Collection, covering indoor and outdoor cameras, a video doorbell, thermostats and door locks.

Ben Wood, chief analyst at CCS Insight, said that despite its “cautious approach”, Apple had been waiting for the upgraded Siri to launch its home range, “given how important the voice interactions are with this product”. He also pointed to the partnerships as a way to reduce installation issues, saying they address “one of the category’s biggest barriers: the complexity of getting things working”.

Meanwhile, analysts at ABI Research said the launches represent “a more serious push into the smart home” by Apple but noted it “still trails Amazon and Google on smart home maturity”, arguing tighter integration with other devices and a personalised user experience will be central to its prospects. They also stated, “Apple’s long-term position will depend on how open it becomes to third-party devices, services, and possibly external AI agents.”

Apple revamps smart home strategy with HomeView

AT&T strikes $2B sale to exit Mexico

What happened: AT&T agreed to sell its Mexican business to Grupo Televisa and Fintech Holdings for $2 billion, ending its decade-long attempt to challenge America Movil’s dominance in the country.

Why it matters: Grupo Televisa and Fintech Holdings will each take an equal stake in the operation. AT&T Mexico had 24.7 million wireless subscribers as of 31 December 2025, equivalent to around 16% of the market. Meanwhile, incumbent America Movil’s Telcel held approximately 58%.

LightShed Partners analysts Walter Piecyk and Joe Galone described the move as part of a wider strategic shift under CEO John Stankey, noting AT&T is increasingly focused on “US wireless and highly durable fibre business”. They said the $2 billion sale proceeds represent less than five weeks of AT&T’s annual capital investment, arguing the Mexican operation has limited relevance to its future priorities. They also pointed to America Movil’s resilience, noting that “Telcel still controls nearly 60% of the market, while AT&T and Telefonica are heading for the exits”.

The analysts further highlighted a potential competitive threat from SpaceX, whose combination of terrestrial spectrum and satellite capacity could challenge traditional wireless economics.

AT&T strikes $2B sale to exit Mexico

Nokia offloads Enterprise Campus Edge unit

What happened: Nokia agreed to sell its Enterprise Campus Edge (ECE) business to investment firm Aurelius Group, as it continues to offload assets and streamline its core strategic focus.

Why it matters: ECE serves almost 900 customers across manufacturing, mining, logistics, airports, ports and power generation. Its portfolio spans private wireless networks, industrial devices, on-premises operational technology, edge computing and software for industrial AI applications. Nokia said the sale would allow it to focus capital and resources on areas offering greater long-term value. ECE will operate as a standalone company under Aurelius, with employees expected to transfer when the deal closes. Earlier this month, Nokia sold its fixed wireless access customer premises equipment business to US broadband vendor Inseego.

Future Wireless principal analyst Pablo Tomasi called ECE sale “the end of a bold chapter” in Nokia’s history, arguing the vendor had abandoned its push to become an industrial vendor. Kyung Mun, principal analyst at Mobile Experts noted ECE was “a smaller part of its overall private networks business”, which also includes railway and defence. He noted Nokia held more than 50% of core and RAN equipment sales in 2025 including both industrial and enterprise segments, adding: “The ECE business grew in the low double digits. Let’s see how it does under the new owner.”

Nokia offloads Enterprise Campus Edge unit

The post The Friday File: Apple HomeView; AT&T Mexico; Nokia ECE appeared first on Mobile World Live.

Oppo outlined plans to boost investment in Europe significantly over the next three years, as it seeks to strengthen its position in the premium smartphone market and rival Apple and Samsung in the region.

During a press briefing in London, Oppo Europe CEO Elvis Zhou (pictured) revealed it plans to increase investment by 30% year-on-year up to 2029, however it did not provide a figure on how much it invested in 2025. It has also not publicly disclosed previous investment figures for its European operations.

Zhou was clear on a renewed commitment to Europe, adding it aims to focus on the UK, France, Germany, Italy and Spain as its main markets.

Referring to competitors, Zhou said: “We are not here to say we will replace one of these established brands. No, we are here to offer a differentiated third option.” While acknowledging rivals’ advantages in brand awareness and retail distribution, he explained that capturing between 5% and 10% of Europe’s premium smartphone market was “totally feasible” for Oppo.

He said the vendor is targeting younger consumers, particularly those aged 18 to 35, by putting a spotlight on camera performance, battery life and AI features rather than competing primarily on price. A market survey by Oppo showed that more than 40% of young people weighed AI tools as a deciding factor when purchasing devices.

In Zhou’s view, “Oppo is not a brand shy of investment… not because we have more money; it’s because we know which kind of segment of the clients we want to target, and what kind of products can really meet their demands”.

He added the company also aims to double sales of its high-end Find flagship series in 2027 compared with 2026. According to Omdia figures, more than half of Oppo devices sold in the UK during the first half of 2026 were Find series models. The series’ share of Oppo’s European sales also increased fourfold in 2026 compared with last year, “showing the growing contribution of our premium products to our European business”.

His comments came as Oppo prepares for the global launch of its latest Find X10 and Find X10 Pro Max smartphone models on Tuesday (21 October).

Firm stand
Despite Zhou’s bullish outlook, the broader challenge is persuading consumers to pay more as replacement cycles lengthen. He noted Oppo does not intend to compete by offering cheaper devices.

“If we want to have a firm stand here in the European market, it’s not because we are affordable; it’s because we can offer a differentiated and good product experience,” he said, adding: “Our positioning regarding the European market is clear cut. This is a high-end premium market.”

However, the chief cautioned against expecting rapid gains, stating the company would prioritise steady growth. “We will not resort to short-term discount just to increase the sales volume,” he said. “We might not achieve a lot of short-term successes like impressive market share … but you will see our pace here. It will be really stable.”

Oppo operates across more than 20 countries in Europe and is expanding into the Nordics, Poland and Romania, with Spain currently its largest market in the region. Globally, the vendor claimed it ranked among the four largest smartphone vendors by shipments in the first half of 2026, adding the share of its global shipment value generated overseas exceeded 60%.

The Chinese vendor’s latest push in the continent follows a period of disruption after a 2022 patent dispute with Nokia led to a suspension of its sales in Germany. Subsequently in 2023, Oppo reaffirmed its long-term European ambitions despite the loss of its exclusive distributor in France, Yang Technology SAS, and the continued absence of its products from the German market.

The post Oppo vows to challenge Apple, Samsung Europe duopoly appeared first on Mobile World Live.

Google DeepMind co-founder Demis Hassabis cautioned that recent AI-driven cyberattacks were “early warning shots”, as more serious incidents could emerge over time unless companies and governments stepped up efforts to secure AI systems.

Speaking at the International AI Summit in Dublin, Hassabis said the industry needs to adopt safety practices used in sectors such as biology and nuclear technology, calling for secure, isolated sandboxes to test AI agents before they are connected to the internet, alongside stronger cybersecurity measures.

“We’ve got to kind of asymmetrically power up our defensive side to keep ahead of the attackers,” he said, arguing banks and governments could deploy AI to identify and fix software vulnerabilities before systems are widely released.

He also renewed calls for an international standards body to oversee AI safety, but clarified any framework must be flexible enough to keep pace with technological developments without stifling innovation.

Microsoft president and CEO Brad Smith also advocated for stronger safeguards, describing 2026 as a “sobering year” for AI safety. “No company or technology should be above the law,” he said, adding that regulation needed to be “sensible” and “balanced” while accounting for innovation.

Smith added that companies should be prepared to slow development when necessary, citing Microsoft’s decision to move 34,000 engineers from product development to cybersecurity two and a half years ago. “It was painful, but it was the right thing to do.”

“This allergy that some in the industry have to regulation is frankly something we need to get over,” he quipped.

European AI research
Separately, Google DeepMind and Google Cloud announced plans to provide up to 200 European scientific researchers with access to their frontier AI models, alongside $3 million in compute credits.

The initiative will support the European Union’s AI for Science agenda through the Resource for AI Science in Europe (RAISE), which aims to address computing resource gaps faced by researchers.

The companies will provide researchers with access to AI tools including AlphaFold 3, WeatherNext, AlphaGenome, AlphaEvolve and AlphaEarth Foundations. Google said the programme would equip researchers to tackle complex scientific challenges, including predicting extreme weather patterns and mapping molecular structures.

The post AI chiefs sound alarm on security risks appeared first on Mobile World Live.

NTT Docomo and US-based network equipment manufacturer Abside Networks signed an agreement to explore next-generation, AI-driven networks using high-altitude platform stations (HAPS), as operators look to extend networks beyond terrestrial infrastructure.

Under the memorandum of understanding (MoU), the pair will collaborate on HAPS system architecture, network control and data utilisation. They will also examine how AI can forecast demand and manage radio beams, capacity, spectrum and interference in stratospheric RAN.

In addition, the partners plan to study the collection and use of data on network quality, operations and stratospheric sensing, as well as potential integration with external applications and service platforms.

The agreement supports NTT Docomo’s broader vision for a “three-dimensional communications infrastructure that integrates AI with non-terrestrial network (NTN) initiatives, including HAPS”. Operating at altitudes of around 20km, HAPS use unmanned platforms to provide wide-area connectivity, including to underserved locations where conventional mobile coverage is difficult or uneconomical, such as mountainous regions and maritime areas.

The Japanese operator explained it aims to integrate terrestrial networks with systems in the stratosphere and space to enable wider-area, higher-capacity services that address mounting connectivity demand as AI usage balloons.

NTT Docomo has been conducting research and demonstrations aimed at delivering HAPS-based communications services, but noted the stratospheric environment requires specialised RAN technologies and equipment.

As a result, Abside is developing radio systems based on large antenna arrays for precise beam control, alongside an AI-driven software platform designed to provide “integrated control of networks across the ground, stratosphere and space”.

The equipment manufacturer also brings experience in stratospheric communications through its work on Google’s Project Loon balloon initiative, alongside development of HAPS equipment, NTN technologies and AI-based network control systems.

The post Docomo, Abside team on AI for HAPS communications appeared first on Mobile World Live.

The GSMA board outlined the industry’s connectivity priorities, calling for action on smartphone affordability, online scams and network investment while backing closer international cooperation on future 6G standards.

Following a board meeting in New Delhi, India, the industry association stated that although 97% of the global population live within mobile broadband coverage, 3.4 billion people remain unconnected to mobile internet. The cost of devices remains a significant barrier, with an entry-level internet-enabled handset equivalent to 44% of average monthly income for the poorest 20% of people in low- and middle-income countries.

The GSMA highlighted potential measures to improve affordability including financing schemes, targeted tax cuts and closer cooperation between device makers and memory suppliers.

It also identified scams and fraud as a growing threat, warning that although the mobile industry continues to strengthen network detection and cyber defence tools, “no single sector can tackle the threat alone”. The GSMA noted AI-generated content and deepfakes further add to risks for consumers, public trust and government objectives, calling on the industry to “join forces to counter scams with practical solutions”.

Sustained investment
The GSMA also highlighted the need for network investment and supportive policy to accelerate connectivity growth. It found mobile technologies and services generated $6 trillion in economic value in 2025, equivalent to 6.4% of global GDP. This figure is forecast to reach $11.3 trillion by 2030.

In addition, operators are projected to invest $1.2 trillion in capital expenditure between 2025 and 2030 to expand and upgrade networks, with 57% of mobile connections expected to use 5G by the end of the decade.

GSMA director general Vivek Badrinath said the sector’s next priority must be to translate 5G expansion into “meaningful value”, ensuring more people can participate in the digital economy. “This will require sustained investment, enabling policy and collaboration across the wider ecosystem,” he stressed.

Looking ahead, the industry association reiterated its support for global standards and harmonised spectrum as 6G development continues. It highlighted countries representing more than 80% of the world’s population support mobile use of the upper 6GHz band, which could provide the 200MHz to 400MHz channel bandwidths expected to be needed for 6G.

GSMA chair Gopal Vittal said that “by working through global standards and collaborating across markets, we can accelerate innovation, strengthen trust and resilience” as the industry develops the next generation of connectivity.

The post GSMA puts 6G, AI threats, digital inclusion in focus appeared first on Mobile World Live.

Meta Platforms CEO Mark Zuckerberg (pictured) reportedly claimed the company had reached the “frontier” of AI, as it unveiled two new models designed to handle complex personal and professional tasks.

Following hefty investments in computing infrastructure and research, the technology giant said Muse Wonder and Muse Wonder Full AI mark “our most capable models to date”.

The models can handle tasks ranging from coding and document analysis to interpreting legal information, personal finance and health data.

Among key features is Active Memory, enabling the retention of relevant user information across conversations. The agents are also equipped with Active Focus to help them continue pursuing longer tasks, adapt to new information, recover from errors and ask users for clarification before taking important actions.

Meta claimed its evaluations showed strong results across benchmarks covering agent performance, memory and coding, adding the advanced Muse Wonder Full version performs better on unfamiliar tasks.

Built responsibly
Unlike its earlier open-source strategy, Meta is offering Muse Wonder through its own developer and subscription channels in a direct challenge to AI chatbots from rivals including OpenAI and Anthropic. According to the Financial Times (FT), Meta said it would charge companies from between 20 cents per million tokens for basic access to as much as $200 a month for the more powerful Muse Wonder Full. 

Speaking to the publication on the launch in which he made the comment about frontier AI, Zuckerberg added the new models would enable “much more sophisticated behaviours that are at the frontier of what is being done in the AI field”.

Meta claims Muse Wonder surpasses Anthropic’s Claude Opus 5.5 and Fable 5.1 and OpenAI’s GPT-6 Astra on certain technical measures.

Zuckerberg added that reaching this point required “a large amount of compute and investment” alongside sustained research. He told the FT that further model development and spending was under way, stating the company was “already working on the next step up”.

The chief also played down mounting concerns about increasingly capable AI systems escaping human control, declaring: “I’m much closer to the belief that AI is a technology that can be engineered with normal guardrails and built responsibly.”

Meta said its new models were trained to prevent unauthorised disclosure of private information and reject requests to facilitate harmful actions, while retaining agentic capabilities for defensive cybersecurity work.

The post Meta boss takes AI fight to rivals with Muse launches appeared first on Mobile World Live.

Elon Musk pledged to make Starlink affordable in India as he renewed his attack on billionaire Mukesh Ambani’s (pictured) telecoms dominance, days before Jio Platforms launches a record-breaking IPO.

In a series of posts on social media platform X, the SpaceX chief argued that competition from Starlink would benefit Indian consumers by expanding connectivity options. “The reality is that competition would greatly benefit the average citizen in India by giving them more choices, as it always does,” he explained, adding, “Ambani’s profits will be lower, but he is not going out of business”.

Musk also dismissed claims that Starlink would struggle to compete with Jio’s low-cost data plans while also posing a threat to the incumbent operator. “This is absurd hypocrisy and hyperbole,” he said.

His comments come as Jio Platforms targets a valuation of around $106 billion, according to Reuters, in a planned IPO that could become India’s largest-ever listing. The business includes telecoms operation Reliance Jio, which reported 524.4 million subscribers at the end of March, including 268.5 million 5G customers.

“Prime Minister Ambani“
Musk has escalated his criticism of India’s regulatory process in recent days, accusing authorities and “certain oligarchs” of delaying Starlink’s entry into the country to maintain a “monopolistic chokehold on the Indian people”.

In a separate post on Friday (9 October), he sarcastically addressed the Jio Platforms-owner as “Prime Minister Ambani” and asked whether he would “consider allowing Starlink to compete”.

He argued satellite connectivity could bring internet access to underserved communities, improve educational opportunities and enable small businesses to reach international markets.

“Please accept my humble apologies for not realizing that you are the real boss of India,” he wrote, adding “Starlink would change their lives for the better”.

India’s Ministry of Communications has rejected allegations of discriminatory treatment, stating that Starlink and incumbents Bharti Airtel and Reliance Jio were at broadly the same stage of the regulatory process. All three have secured initial approvals but remain subject to outstanding security clearances before they can obtain spectrum access.

The post Musk takes fresh swipe at Ambani as Jio IPO looms appeared first on Mobile World Live.

Nvidia is reportedly in talks to either acquire or increase its investment in open-source start-up Reflection AI, as it seeks to strengthen its position in open-weight AI models and challenge lower cost Chinese rivals.

Sources told the Financial Times (FT) discussions remain at an early stage and could lead to a full acquisition, further equity investment or a deal covering chips and computing capacity. An acqui-hire, under which Nvidia would recruit staff and license technology, is another option.

A deal could apparently be reached in the coming weeks, though the terms under discussion remain unclear.

According to the report, Nvidia has already invested $800 million in Reflection AI, which was valued at $25 billion in a March funding round. Founded in 2024 by former Google DeepMind researchers, the start-up develops AI models and tools for software development.

Reflection launched its first open-weight model, Beam, last week, targeting coding and agentic tasks as it looks to compete with Chinese models such as DeepSeek and Kimi.

Open-weight models allow businesses and developers to customise AI systems and run them on their own infrastructure, potentially cutting costs compared with closed services such as OpenAI’s ChatGPT and Anthropic’s Claude.

Although Nvidia has its own Nemotron family of open-weight models, FT said it lacks a frontier-scale model of its own. In an essay published in July, CEO Jensen Huang called for a domestic open-weight “ecosystem” across the US, arguing it was essential for “creating opportunities for innovation and prosperity across the country”. “In fact, openness may be one of the most important paths to AI safety and security,” he argued, adding “relying solely on closed models is not inherently safe” as “they can be breached, misused, or fail in ways that outsiders cannot detect”.

Reflection AI has also secured substantial computing capacity from SpaceX. The deal, struck in June, gives the start-up access to Nvidia’s GB300 chips at SpaceX’s Colossus 2 facility in Memphis, Tennessee, for $150 million per month through 2029.

The post Nvidia weighs Reflection AI acquisition appeared first on Mobile World Live.

Mobile World Live (MWL) brings you our top three picks of the week as Indian authorities rejected Elon Musk’s allegations about Starlink’s regulatory roadblocks, BT rescued TalkTalk in a £400 million deal and Deutsche Telekom (DT) looked to capitalise on sovereign AI demand.

India hits back at Musk over Starlink oligarch claims

What happened: India’s Ministry of Communications rebuked Elon Musk’s accusation that Indian oligarchs were blocking Starlink’s launch in the country and seeking to preserve a “monopolistic chokehold on the Indian people”, stressing its satellite communications framework is “fair and non-discriminatory”.

Why it matters: The watchdog explained Starlink, along with Indian telecom incumbents Reliance Jio and Bharti Airtel, are all at the same stage of the regulatory process. Each operator must complete security assessments before they can seek spectrum, with final clearances pending. Musk has previously clashed with Reliance Jio owner Mukesh Ambani over satellite spectrum access in the country. Starlink’s protracted approval process in India has lasted four years, with security requirements and spectrum access among key issues. Its presence in the country has also been complicated by reports that Starlink equipment was recovered across parts of the country, claims Musk said do not indicate an illegal active service in India.

Manas Mody, founder of business consultancy FX & Float, said the regulatory debate was “broader than a contest between Musk and operators” because “there could be legitimate regulatory and security considerations behind the resistance”, adding India “has done considerable work to secure its critical digital infrastructure”. “While Musk’s concerns may have some merit, I believe there’s likely more to this than just a conspiracy involving oligarchs,” he argued, adding Starlink is “unlikely to make a significant dent in India’s consumer telecom market” at its planned pricing as its “real opportunity lies in improving connectivity in underserved regions and remote hubs”.

India hits back at Musk over Starlink oligarch claims

BT strikes £400M deal for TalkTalk

What happened: BT Group agreed a £400 million deal to acquire TalkTalk’s consumer and wholesale businesses out of administration, bringing 2.5 million customers and around 900 employees under its ownership.

Why it matters: The acquisition prevents the UK broadband provider from folding and potentially disrupting services for millions of customers. TalkTalk reported £1.2 billion in revenue in the past 12 months but was loss-making following a prolonged, unsuccessful sales process. BT CEO Allison Kirkby called the situation “genuinely unprecedented”, warning millions of citizens, businesses and public services were at risk if TalkTalk collapsed. She said the acquisition was “the only viable option to keep millions of customers connected and supported”, with the BT adding the business can become value accretive if it is stabilised.

Kester Mann, director of consumer and connectivity at FDM CCS Insight, said the deal “will inevitably also evoke scrutiny from regulators keen to ensure a fair and competitive playing field”, given BT’s “strong position in the broadband market”. He added that this was particularly relevant after competition authorities raised concerns over Nexfibre’s planned acquisition of fellow broadband player Netomnia.

For TalkTalk, Mann said the deal “cements a painful decline” as the company “failed to respond sufficiently to a fast-changing and increasingly competitive market”. Its demise, he warned, should be “a stark warning to any value-focused telecom provider”.

BT strikes £400M deal for TalkTalk

DT eyes sovereign AI revenue bump

What happened: DT set out plans to turn rising demand for sovereign AI into a new revenue stream, targeting around €250 million in non-US AI-related revenue this year and approximately €800 million by 2030.

Why it matters: The German operator noted European businesses and public institutions seeking AI capabilities while retaining control of data and operations presented a key opportunity, describing sovereign AI as a “growth business”.To that end, DT said it is developing an automation platform for SMEs and expanding AI customer service offerings. CEO Timotheus Hoettges stressed it is crucial AI “delivers measurable value” across the business.

Jamie Escott, account director at Vodafone Business Security Enhanced, said DT’s projected €2.5 billion reduction in indirect costs by 2030 highlights “the growing importance of AI moving beyond experimentation and into large-scale operational deployment”. He added organisations are increasingly focusing on “measurable outcomes such as productivity, efficiency and service quality”. Meanwhile, Jean-Philippe Romieu, executive director for cloud and AI professional services at Microsoft Germany, described DT’s strategy as a “telco blueprint”, pointing to use cases included network integrated AI, autonomous infrastructure and AI-assisted operations. However, he flagged issues including volatile compute costs and market adoption, warning DT’s success “relies heavily on European enterprises prioritising data localisation over hyperscaler pricing and performance”.

DT eyes sovereign AI revenue bump

The post The Friday File: Starlink; TalkTalk; DT appeared first on Mobile World Live.

Indian authorities pushed back against Elon Musk’s claims that unnamed oligarchs were blocking Starlink from launching in the country, stating all three prospective operators remain broadly at the same stage of the regulatory process.

In a post on X, Musk accused “certain oligarchs” of trying to maintain a “monopolistic chokehold on the Indian people”, adding “you can guess who they are…” without naming the parties involved. “This is a crime against the people of India!” he wrote.

In a statement, India’s Ministry of Communications rejected the allegation, explaining its satellite communications framework is “fair and non-discriminatory”. It said assessments are currently underway for all licensed operators in India, with each required to meet security conditions before it can seek spectrum.

“At present, all the three entities are at broadly the same regulatory stage,” the ministry said, calling claims of discriminatory treatment “baseless and misconceived”.

The country’s telecoms sector is dominated by Mukesh Ambani’s Reliance Jio and Sunil Mittal’s Bharti Airtel. Both operators, alongside Musk’s Starlink, have secured initial approvals for satellite services but are awaiting spectrum assignment and final clearances.

Starlink’s approval process has stretched over four years, with security requirements and spectrum access among the key issues under scrutiny. Musk has previously clashed with Ambani over India’s satellite spectrum allocation, although both Reliance Jio and Bharti Airtel later struck agreements with the tech mogul to distribute Starlink services in the country.

The US satellite venture’s regulatory wait has also been marred by security concerns. Local media reports stated authorities recovered illegal Starlink equipment in two separate operations in late 2024, including a drug seizure in the Andaman and Nicobar Islands and an army search in the northeastern state of Manipur.

This week, Musk also disputed any suggestion the equipment was connected to an active Starlink service in India, posting on X that its satellite beams were “turned off over India” and “never on in the first place”.

The post India hits back at Musk over Starlink oligarch claims appeared first on Mobile World Live.

US non-profit Common Sense Media’s Youth AI Safety Institute declared OpenAI’s ChatGPT for Teens an “unacceptable risk” and called for the company to halt access for users aged under 18-years until safeguards are independently tested and improved.

The institute tested more than 4,000 prompts across ChatGPT accounts registered to users aged 13 to 17 years-old before and after the teen-focused service launched in August.

Although some safeguards including the chatbot’s refusal to engage in explicit sexual roleplay functioned as expected, several other protections failed to operate.

“We call on OpenAI to pause marketing the product and to keep teens off of ChatGPT until it can offer a safe, developmentally appropriate experience,” the institute stated.

The biggest concern were parental notifications. In dedicated tests, parent-linked accounts generated no alerts during conversations involving high-risk topics including suicide, self-harm or eating disorders, including sessions lasting up to an hour. Broader testing triggered four notifications which only appeared on older accounts with weeks of sensitive history, with some arriving hours or days after crisis-level disclosures.

In addition, ChatGPT for Teens did not reliably refer high-risk accounts to a crisis hotline or professional help. The AI chatbot missed more than one-in-four situations where a crisis referral was warranted, falling below its 95% threshold in three-of-five severe harm categories.

The institute further claimed OpenAI’s age estimation technology repeatedly failed to shift adult-registered test accounts into the teen experience. It also found ChatGPT continued to use language which “simulates preferences and feelings, validates and engages, and continues engagement”, despite the AI company’s updated guidelines for those aged under 18-years.

Mental health
The AI player introduced parental controls and safety notifications to ChatGPT in 2025, stating its systems could identify potential signs of self-harm and alert parents in cases of acute distress.

ChatGPT For Teens emphasised learning, age-appropriate safeguards and parental controls designed to limit exposure to sensitive content.

OpenAI has been hit with a number of lawsuits from parents across the US alleging its chatbot encouraged at-risk teenagers to engage in self-harm, suicide or violent crimes.

In 2025, the US Federal Trade Commission reportedly launched an investigation into OpenAI, Meta Platforms and Character.AI over concerns chatbots were affecting children’s mental health.

The post ChatGPT for Teens deemed to pose unacceptable risk appeared first on Mobile World Live.

Elon Musk (pictured) ruled out Taiwan Semiconductor Manufacturing Company (TSMC) taking an operational role in Terafab, the billionaire’s advanced semiconductor manufacturing facility in the US state of Texas.

The tech mogul explained TSMC’s only potential involvement under discussion is leasing part of the planned site. “Maybe TSMC subleases part of the Terafab if they want, but nothing more than that,” he said in a post on X. He added: “No, we will build and run the fab. Let there be zero doubt about that.”

The comments clarify Musk’s earlier remarks on X that he was in discussions with TSMC and that “something may come of it”, which fuelled speculation that the Taiwanese chipmaker could become involved in operating the facility being developed by SpaceX and Tesla.

Meanwhile, Intel CEO Lip-Bu Tan told Bloomberg the company will continue working with Musk on Terafab, following questions over Intel’s role in the project after Musk disclosed discussions with TSMC. The chipmaker signed up as a development partner in April, working with SpaceX, Tesla and xAI on silicon fab technology for the project.

SpaceX and Tesla announced the Terafab project in August, stating it would combine chip manufacturing, advanced packaging and testing for logic and memory devices at a site projected to exceed 100 million square feet. The facility will produce chips for Musk’s companies, including processors for Tesla’s humanoid Optimus robots and self-driving Cybercabs, alongside high-power chips for SpaceX’s ambitious orbital AI data centres.

The post Musk plays down TSMC presence at AI fab facility appeared first on Mobile World Live.

Amazon committed more than $1 billion over the next five years to support communities hosting its data centres in a bid to address mounting opposition to AI infrastructure.

Amazon Web Services CEO Matt Garman announced the investment in an essay defending data centre development, arguing the buildout is needed for the US to remain competitive in the global AI race.

“Data centres bring considerable tax revenue, positive economic impact, large numbers of good jobs, education programs and other benefits to the communities where they’re built,” he wrote.

The new programme will fund education, workforce training, local hiring, energy upgrades, water and energy preservation projects and grants for local initiatives. Amazon added local communities will help determine how the funding is used.

The company said it expects to connect more than 300,000 students to free degree programmes in data centre communities over five years. It also plans to expand its network of training centres, alongside grants for energy efficiency upgrades targeting more than 300 schools and community buildings and over 30,000 homes.

Amazon also pledged to stop using nondisclosure agreements with government agencies involved in its data centre projects, explaining it would publish annual information on energy consumption, water use and efficiency.

Losing ticket
Data centre expansion has been facing growing backlash across the world over power demand, water consumption, land disruption, noise and environmental impact. In July, New York became the first US state to introduce a one-year moratorium on the construction of hyperscale facilities requiring 50 megawatts of power or more, citing environmental concerns, pressure on the energy grid and costs for residents.

Garman warned data centre moratoriums could leave the country behind in the global AI race. “If these measures are enacted, the US could be writing its own losing ticket to this race, and the consequences would last generations,” he said.

He also pushed back against concerns over water consumption, electricity prices and emissions from backup generators, describing several commonly cited claims as “myths”, “misinformation and outright lies”. He explained direct data centre water consumption accounts for about 0.5% of total US industrial water use and that backup generators remain idle 99.9% of the time.

The post Amazon pledges $1B to ease data centre backlash appeared first on Mobile World Live.

Mobile World Live brings you our top three picks of the week as European Union (EU) governments pushed for changes to high-risk supplier rules, Nvidia’s chief weighed in on the AI model distillation debate and Poste Italiane boosted its stake in Telecom Italia.

Governments press EU to change high-risk supplier plan

What happened: European governments pushed for amendments to a European Commission (EC) proposal that would require operators to remove network equipment supplied by vendors deemed high risk, according to a draft document seen by Reuters.

Why it matters: Early this year, the EC published a Cybersecurity Act proposal outlining a 36-month phase out for high-risk third-country supplier equipment in mobile networks. However, EU member states have reportedly called for the phase out period to be determined by several factors rather than a fixed deadline, including level of risk, equipment lifespans, compatibility requirements and the availability of alternatives. The proposed rules are expected to affect Chinese vendors Huawei and ZTE.

Mason Hurlocker of Bloomsbury Intelligence and Security Institute (BISI) stated “the reforms carry a range of political, operational, security and economic risks”, adding it is possible the EC narrows its provisions “given industry cost warnings and resistance from some member states to mandatory rip-and-replace obligations”. A GSMA Intelligence report estimated mobile network replacement would cost the industry between €30 billion and €40 billion.

Governments press EU to change high-risk supplier plan

Nvidia chief says AI model distillation is fair competition

What happened: In an interview with CNBC, Nvidia CEO Jensen Huang said AI model distillation should be treated as competition rather than theft, putting him at odds with US officials over the practice.

Why it matters: Huang said developers are entitled to test competing products, comparing distillation with competitors taking Nvidia chips apart to understand how they work. “That’s called competition,” he apparently said. He further argued the burden is on providers to control access, stating: “if you don’t like people to use your products… disable the service.”

The comments come after US officials accused Chinese AI players of conducting industrial-scale distillation campaigns against US models. Anthropic separately alleged that Chinese companies including Alibaba and DeepSeek used large numbers of accounts and proxy networks to extract capabilities from Claude.

Otis Yi, programme director at digital entertainment platform GashPoint, said “the dispute centres on how those outputs are obtained and whether the model owner has authorised their use”, adding industrial-scale extraction can raise “separate questions about contracts, intellectual property and enforcement”. Mark Seery, founder of consulting and technology advisory firm Bohcay, argued: “Competition matters. So does authorisation.” He noted “the customer does not own the underlying infrastructure”, adding when they exceed agreed rights, “that is not simply “testing a product”; it may be a contractual violation with legal consequences”.

Nvidia chief says AI model distillation is fair competition

Poste Italiane secures 85% stake in Telecom Italia

What happened: Poste Italiane secured an 85.8% stake in Telecom Italia (TIM) after completing the second and final phase of its public tender offer.

Why it matters: The postal group acquired an additional 65.7% of TIM, building on the 20% stake it acquired last year when it replaced Vivendi as the operator’s largest shareholder. The tender offer, launched in July, valued the transaction at around €13 billion and offered shareholders €1.67 in cash per share.

In March, the company had outlined plans to take TIM private. However, Reuters reported the result leaves Poste well short of the 90% threshold needed to take it private, leaving it to consider other options.

The operator’s board has since acknowledged Poste’s exercise of “direction and coordination” over TIM and its subsidiaries, and approved new rules covering operational governance and information flows between the companies. Alessandro Marchesini, a TIM board member, described the transaction as the closing of an “important financial chapter”, adding, “now the industrial one begins”. He said the focus would shift to “turning the opportunities created by this transaction into projects, collaboration and long-term value”.

Poste Italiane secures 85% stake in Telecom Italia

The post The Friday File: High risk vendors; Nvidia; TIM appeared first on Mobile World Live.