Internet of things

Privacy-preserving AI gets a boost with Google’s VaultGemma model

Google has unveiled VaultGemma, a new large language model built to offer cutting-edge privacy through differential privacy. The 1-billion-parameter model is based on Google’s Gemma architecture and is described as the most powerful differentially private LLM to date.

Differential privacy adds mathematical noise to data, preventing the identification of individuals while still producing accurate overall results. The method has long been used in regulated industries, but has been challenging to apply to large language models without compromising performance.

VaultGemma is designed to eliminate that trade-off. Google states that the model can be trained and deployed with differential privacy enabled, while maintaining comparable stability and efficiency to non-private LLMs.

This breakthrough could have significant implications for developers building privacy-sensitive AI systems, ranging from healthcare and finance to government services. It demonstrates that sensitive data can be protected without sacrificing speed or accuracy.

Google’s research teams say the model will be released with open-source tools to help others adopt privacy-preserving techniques. The move comes amid rising regulatory and public scrutiny over how AI systems handle personal data.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

EdChat AI app set for South Australian schools amid calls for careful use

South Australian public schools will soon gain access to EdChat, a ChatGPT-style app developed by Microsoft in partnership with the state government. Education Minister Blair Boyer said the tool will roll out next term across public high schools following a successful trial.

Safeguards have been built into EdChat to protect student data and alert moderators if students type concerning prompts, such as those related to self-harm or other sensitive topics. Boyer said student mental health was a priority during the design phase.

Teachers report that students use EdChat to clarify instructions, get maths solutions explained, and quiz themselves on exam topics. Adelaide Botanic High School principal Sarah Chambers described it as an ‘education equaliser’ that provides students with access to support throughout the day.

While many educators in Australia welcome the rollout, experts warn against overreliance on AI tools. Toby Walsh of UNSW said students must still learn how to write essays and think critically, while others noted that AI could actually encourage deeper questioning and analysis.

RMIT computing expert Michael Cowling said generative AI can strengthen critical thinking when used for brainstorming and refining ideas. He emphasised that students must learn to critically evaluate AI output and utilise the technology as a tool, rather than a substitute for learning.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

YouTube expands AI dubbing to millions of creators

Real-time translation is becoming a standard feature across consumer tech, with Samsung, Google, and Apple all introducing new tools. Apple’s recently announced Live Translation on AirPods demonstrates the utility of such features, particularly for travellers.

YouTube has joined the trend, expanding its multi-language audio feature to millions of creators worldwide. The tool enables creators to add dubbed audio tracks in multiple languages, powered by Google’s Gemini AI, replicating tone and emotion.

The feature was first tested with creators like MrBeast, Mark Rober, and Jamie Oliver. YouTube reports that Jamie Oliver’s channel saw its views triple, while over 25% of the watch time came from non-primary languages.

Mark Rober’s channel now supports more than 30 languages per video, helping creators reach audiences far beyond their native markets. YouTube states that this expansion should make content more accessible to global viewers and increase overall engagement.

Subtitles will still be vital for people with hearing difficulties, but AI-powered dubbing could reduce reliance on them for language translation. For creators, it marks a significant step towards making content truly global.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Cyberattack keeps JLR factories shut, hackers claim responsibility

Jaguar Land Rover (JLR) has confirmed that data was affected in a cyberattack that has kept its UK factories idle for more than a week. The company stated that it is contacting anyone whose data was involved, although it did not clarify whether the breach affected customers, suppliers, or internal systems.

JLR reported the incident to the Information Commissioner’s Office and immediately shut down IT systems to limit damage. Production at Midlands and Merseyside sites has been halted until at least Thursday, with staff instructed not to return before next week.

The disruption has also hit suppliers and retailers, with garages struggling to order spare parts and dealers facing delays registering vehicles. JLR said it is working around the clock to restore operations in a safe and controlled way, though the process is complex.

Responsibility for the hack has been claimed by Scattered Lapsus$ Hunters, a group linked to previous attacks on Marks & Spencer, the Co-op, and Las Vegas casinos in the UK and the US. The hackers posted alleged screenshots from JLR’s internal systems on Telegram last week.

Cybersecurity experts say the group’s claim that ransomware was deployed raises questions, as it appears to have severed ties with Russian ransomware gangs. Analysts suggest the hackers may have only stolen data or are building their own ransomware infrastructure.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Arm unveils Lumex mobile chips built for AI

UK-based chip designer Arm introduced Lumex, a next-generation chip design explicitly designed to power AI on smartphones, smartwatches, and next-generation PCs.

Arm, whose processor architecture underpins devices from Apple and Nvidia, described Lumex as its most advanced platform yet for real-time AI assistance, communication and on-device personalisation.

Arm’s senior vice-president Chris Bergey said consumers now expect instant, private, seamless AI features instead of gradual improvements.

The Lumex platform combines performance, privacy, and efficiency, allowing partners to use the design as delivered or configure it to their own requirements.

A brand that is part of a broader naming structure includes Neoverse for infrastructure, Niva for PCs, Zena for automotive, and Orbis for the internet of things.

Meanwhile, Arm is reportedly preparing to manufacture its chips, having recruited Amazon’s Rami Sinno, who helped build Trainium and Inferentia, to strengthen its in-house ambitions.

These moves mark a significant moment for Arm, as the company seeks to expand its influence in the AI hardware space and reduce reliance on licensing alone.

With the rise of generative AI, the push for high-performance chips designed around on-device intelligence is becoming central to the future of mobile technology.

Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!

AI in weather forecasting takes centre stage in Abu Dhabi

Abu Dhabi hosted a Weather Summit that explored how AI could transform forecasting and support operations, such as cloud seeding. Experts emphasised that AI enhances analysis but must complement, rather than replace, human judgement.

Discussions focused on Earth-system forecasting using satellite datasets, IoT devices, and geospatial systems. Quality, interoperability, and equitable access to weather services were highlighted as pressing priorities.

Speakers raised questions about public and private sector incentives’ reliability, transparency, and influence on AI. Collaboration across sectors was crucial to strengthening trust and global cooperation in meteorology.

WMO President Dr Abdulla Al Mandous said forecasting has evolved from traditional observation to supercomputing and AI. He argued that integrating models with AI could deliver more precise local forecasts for agriculture, aviation, and disaster management.

The summit brought together leaders from UN bodies, research institutions, and tech firms, including Google, Microsoft, and NVIDIA. Attendees highlighted the need to bridge data gaps, particularly in developing regions, to confront rising climate challenges.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Social media authenticity questioned as Altman points to bot-like behaviour

Sam Altman, X enthusiast and Reddit shareholder, has expressed doubts over whether social media content can still be distinguished from bot activity. His remarks followed an influx of praise for OpenAI Codex on Reddit, where users questioned whether such posts were genuine.

Altman noted that humans are increasingly adopting quirks of AI-generated language, blurring the line between authentic and synthetic speech. He also pointed to factors such as social media optimisation for engagement and astroturfing campaigns, which amplify suspicions of fakery.

The comments follow OpenAI’s backlash over the rollout of GPT-5, which saw Reddit communities shift from celebratory to critical. Altman acknowledged flaws in a Reddit AMA, but the fallout left lasting scepticism and lower enthusiasm among AI users.

Underlying this debate is the wider reality that bots dominate much of the online environment. Imperva estimates that more than half of 2024’s internet traffic was non-human, while X’s own Grok chatbot admitted to hundreds of millions of bots on the platform.

Some observers suggest Altman’s comments may foreshadow an OpenAI-backed social media venture. Whether such a project could avoid the same bot-related challenges remains uncertain, with research suggesting that even bot-only networks eventually create echo chambers of their own.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Fake GitHub downloads deliver GPUGate malware to EU IT staff

A malvertising campaign is targeting IT workers in the EU with fake GitHub Desktop installers, according to Arctic Wolf. The goal is to steal credentials, deploy ransomware, and infiltrate sensitive systems. The operation has reportedly been active for over six months.

Attackers used malicious Google Ads that redirected users to doctored GitHub repositories. Modified README files mimicked genuine download pages but linked to a lookalike domain. MacOS users received the AMOS Stealer, while Windows victims downloaded bloated installers hiding malware.

The Windows malware evaded detection using GPU-based checks, refusing to run in sandboxes that lacked real graphics drivers. On genuine machines, it copied itself to %APPDATA%, sought elevated privileges, and altered Defender settings. Analysts dubbed the technique GPUGate.

The payload persisted by creating privileged tasks and sideloading malicious DLLs into legitimate executables. Its modular system could download extra malware tailored to each victim. The campaign was geo-fenced to EU targets and relied on redundant command servers.

Researchers warn that IT staff are prime targets due to their access to codebases and credentials. With the campaign still active, Arctic Wolf has published indicators of compromise, Yara rules, and security advice to mitigate the GPUGate threat.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

Conti and LockBit dominate ransomware landscape with record attacks

Ransomware groups have evolved into billion-dollar operations targeting critical infrastructure across multiple countries, employing increasingly sophisticated extortion schemes. Between 2020 and 2022, more than 865 documented attacks were recorded across Australia, Canada, New Zealand, and the UK.

Criminals have escalated from simple encryption to double and triple extortion, threatening to leak stolen data as added leverage. Attack vectors include phishing, botnets, and unpatched flaws. Once inside, attackers use stealthy tools to persist and spread.

BlackSuit, formerly known as Conti, led with 141 attacks, followed by LockBit’s 129, according to data from the Australian Institute of Criminology. Ransomware-as-a-Service groups hit higher volumes by splitting developers from affiliates handling breaches and negotiations.

Industrial targets bore the brunt, with 239 attacks on manufacturing and building products. The consumer goods, real estate, financial services, and technology sectors also featured prominently. Analysts note that industrial firms are often pressured into quick ransom payments to restore production.

Experts warn that today’s ransomware combines military-grade encryption with advanced reconnaissance and backup targeting, raising the stakes for defenders. The scale of activity underscores how resilient these groups remain, adapting rapidly to law enforcement crackdowns and shifting market opportunities.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!

ITU warns global Internet access by 2030 could cost nearly USD 2.8 trillion

Universal Internet connectivity by 2030 could cost up to $2.8 trillion, according to the International Telecommunication Union (ITU) and Saudi Arabia’s Communications, Space, and Technology (CST) Commission. The blueprint urges global cooperation to connect the one-third of humanity still offline.

The largest share, up to $1.7 trillion, would be allocated to expanding broadband through fibre, wireless, and satellite networks. Nearly $1 trillion is needed for affordability measures, alongside $152 billion for digital skills programmes.

ITU Secretary-General Doreen Bogdan-Martin emphasised that connectivity is essential for access to education, employment, and vital services. She noted the stark divide between high-income countries, where 93% of people are online, and low-income states, where only 27% use the Internet.

The study shows costs have risen fivefold since ITU’s 2020 Connecting Humanity report, reflecting both higher demand and widening divides. Haytham Al-Ohali from Saudi Arabia said the figures underscore the urgency of investment and knowledge sharing to achieve meaningful connectivity.

The report recommends new business models and stronger cooperation between governments, industry, and civil society. Proposed measures include using schools as Internet gateways, boosting Africa’s energy infrastructure, and improving localised data collection to accelerate digital inclusion.

Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!