Development Archives | Page 51 of 557 | Digital Watch Observatory

GitHub malware campaign uses SEO tricks to steal browser data

Cybersecurity researchers have uncovered a malware campaign spreading through over 100 GitHub repositories disguised as free software tools. Hackers used SEO-heavy descriptions to make their fake repositories appear high in search results, close to legitimate software.

Users searching for popular programs were directed to counterfeit download pages. These pages offered ZIP files containing BoryptGrab, a malware designed to steal data from infected Windows systems. The files were disguised as cracked software, gaming cheats, or utility tools.

The malware collects sensitive information, including browser passwords, cookies, and cryptocurrency wallet details. It can access nine major browsers, including Chrome, Edge, Firefox, Opera, Brave, and Vivaldi, and bypass some security protections.

Certain variants also install additional tools allowing remote access and persistent control over infected machines. However, this enables hackers to run commands, maintain ongoing access, and steal more information without the user’s knowledge.

Trend Micro, the cybersecurity firm that reported the campaign, noted some code and logs suggest a possible Russian origin, though attribution is not confirmed. Experts warn that GitHub and search engine manipulation make this attack method especially dangerous.

Users are advised to download software only from trusted sources and to verify the authenticity of the repository. Organisations should follow security best practices such as software allowlisting, maintaining inventory, and removing unauthorised applications to prevent similar attacks.

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

AI infrastructure raises critical questions for global technology development

AI is increasingly viewed as a key global infrastructure. The CEO of Nvidia argues that AI should not be seen merely as software but as a foundational technology shaping economies and industries. As a result, companies and governments worldwide are expected to build and rely on AI systems increasingly.

At the same time, AI infrastructure expansion is still in its early stages. Nvidia’s CEO notes that although hundreds of billions of dollars have already been invested in data centres and computing systems, the broader AI buildout will likely require trillions of dollars in additional investment.

Moreover, governance and access decisions will play a critical role. According to Nvidia’s CEO, choices about how quickly AI is developed, who can access it, and how it is regulated will ultimately shape the technology’s long-term impact on society.

In addition, AI differs fundamentally from traditional software. While conventional software follows prewritten instructions, AI systems generate responses dynamically based on context. Consequently, AI can produce new outputs rather than simply retrieving stored commands.

Furthermore, AI development depends on multiple interconnected technological layers. The CEO of Nvidia describes a five-layer stack composed of energy, chips, infrastructure, models, and applications. Each layer supports the next, meaning AI services rely on everything from electricity supply to advanced computing hardware.

Finally, AI may also reshape the labour market. Nvidia’s CEO suggests that as AI increases productivity, companies could expand operations and create new jobs, particularly in infrastructure development and technical fields.

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

EU launches AI platform to detect food fraud and contamination

Food safety monitoring across the EU is receiving a technological upgrade with the launch of TraceMap, a new AI platform designed to detect food fraud, contamination and disease outbreaks more quickly.

The European Commission introduced the tool as part of efforts to strengthen consumer protection and improve oversight of the agri-food supply chain.

TraceMap helps authorities analyse large volumes of data related to food production, distribution and trade. By identifying connections between operators, shipments and supply chains, the system allows investigators to spot suspicious activity and potential safety risks earlier.

National authorities in the EU member states can already access the platform, enabling them to conduct more targeted inspections and investigations without requiring additional resources.

The platform draws on data from existing EU systems such as the Rapid Alert System for Food and Feed (RASFF) and the Trade Control and Expert System (TRACES). Using AI to structure and interpret information, TraceMap can reveal patterns in production and trade flows that may indicate contamination, fraud, or other irregularities in the food supply chain.

Early testing of the platform has already demonstrated its practical value. A pilot version of TraceMap helped authorities identify and recall infant milk formula produced with contaminated ARA oil originating from China.

European officials say the system will strengthen the EU’s ability to respond rapidly to food safety risks while improving monitoring of both domestic production and imported products.

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

Promptfoo joins OpenAI to secure AI deployments

OpenAI is acquiring Promptfoo, a platform designed to help enterprises identify and remediate vulnerabilities in AI systems during development. Once finalised, Promptfoo’s technology will be integrated into OpenAI Frontier, OpenAI’s platform for building and managing AI coworkers.

Promptfoo, led by Ian Webster and Michael D’Angelo, provides tools trusted by over a quarter of Fortune 500 companies. Its open-source CLI and library support evaluation and red-teaming of large language model applications.

The acquisition allows OpenAI to enhance both open-source initiatives and enterprise capabilities within Frontier.

Integration will introduce native security and evaluation features into Frontier. Enterprises will gain automated tools to detect risks such as prompt injections, jailbreaks, data leaks, tool misuse, and out-of-policy agent behaviour.

Security testing will be built into development workflows to catch issues early and support safe AI deployment.

Oversight and accountability features will also be strengthened. Integrated reporting and traceability will allow organisations to document testing, monitor changes over time, and meet governance, risk, and compliance requirements.

The acquisition is expected to expand OpenAI’s ability to deliver secure and reliable AI for enterprise applications.

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

US government faces lawsuits over Anthropic AI move

Anthropic has launched two lawsuits against the US Department of Defence, disputing its recent designation of the AI firm as a ‘supply chain risk.’ The company claims the move is unlawful and infringes on its First Amendment rights.

The company argues that the government is punishing it for refusing to allow the military to use its AI for domestic surveillance or for fully autonomous weapons.

The lawsuits, filed in California and Washington, DC courts, follow the Pentagon’s unprecedented use of the supply chain risk tool against a US company. The designation requires other government contractors to sever ties with Anthropic, posing a serious threat to its business operations.

The company maintains it remains committed to supporting national security applications of its AI.

The Department of Defence has used anthropic’s AI model Claude in operations targeting Iran. The company says it has worked with the DoD on system adaptations and seeks to continue negotiations while protecting its business and partners.

The firm claims government actions cause harm, though CEO Dario Amodei said the designation’s impact is limited. Anthropic insists judicial review is a necessary step to defend its business and ensure the responsible deployment of its technology.

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

Blockchain network Tron joins Agentic AI Foundation to advance AI infrastructure

Tron has joined the Linux Foundation’s Agentic AI Foundation (AAIF) as a governing member to support the development of AI agent infrastructure. The network aims to enable collaboration and interoperability among systems that efficiently manage high-volume, low-value transactions.

Founder Justin Sun highlighted Tron’s speed, scalability, and low fees as key advantages for AI-agent use cases. He noted that as AI agents move to mainstream machine-to-machine commerce, transaction volumes could rise, increasing demand for robust blockchain networks.

The AAIF encourages open-source agentic AI development and establishes standards for governance, safety, and interoperability. Tron joins major members like Circle and JPMorgan while building tools and infrastructure to support AI, including the Bank of AI with AINFT.

Tron currently leads in blockchain revenue, with data showing strong performance over 24 hours, seven days, and 30 days. Sun confirmed that AI activity is contributing to this growth, reflecting the rapid adoption and scaling of agentic AI on the network.

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

Qualcomm and NEURA Robotics partner to accelerate physical AI and cognitive robotics

NEURA Robotics and Qualcomm have formed a long-term strategic collaboration to advance physical AI and next-generation robotics platforms.

A partnership that aims to bring intelligent robots into real-world environments more rapidly by combining advanced AI processors with full-stack robotic systems.

The cooperation focuses on developing ‘Brain + Nervous System’ reference architectures that integrate high-level cognition, such as perception, reasoning and planning, with ultra-low-latency control systems.

Qualcomm’s robotics processors, including the Dragonwing IQ10 Series, will provide AI compute and connectivity, while NEURA contributes robotic hardware platforms and embodied AI software.

Both companies intend to support deployment across multiple robotic forms, including robotic arms, mobile robots, service machines and humanoid platforms.

NEURA’s cloud environment, Neuraverse, will serve as a shared platform for simulation, training and lifecycle management of robotic intelligence, allowing innovations developed by one robot to spread across entire fleets.

The collaboration also aims to establish a global developer ecosystem for robotics applications. Standardised runtime environments and deployment interfaces are expected to simplify how AI workloads move from development into production while maintaining reliability and safety.

Executives from both companies emphasised that robotics represents one of the most demanding AI environments, as decisions must be made instantly and locally.

By combining edge AI processing with cognitive robotic systems, the partnership aims to accelerate commercial deployment of humanoid and general-purpose robots capable of operating safely alongside humans across industries.

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

Microsoft launches Copilot Cowork to automate tasks across Microsoft 365

AI is moving from assistance to execution as Microsoft introduces Copilot Cowork, a system designed to perform tasks across the Microsoft 365 environment.

Instead of simply generating text or suggestions, the feature allows users to delegate real work by describing a desired outcome.

Copilot Cowork converts requests into structured plans that run in the background. The system analyses signals from workplace tools such as Microsoft Outlook, Microsoft Teams and Microsoft Excel to understand schedules, documents and ongoing projects.

Users can approve or modify each step while the AI coordinates actions across meetings, files and messages.

Several enterprise scenarios illustrate the system’s capabilities. Cowork can reorganise calendars by analysing meetings and automatically proposing schedule changes.

It can also prepare complete briefing materials for customer meetings by collecting relevant emails, files and data before generating presentations and research summaries.

The technology also supports deeper analysis tasks. Users can request company research and receive structured outputs that include summaries, financial data and supporting documents.

In product launch planning, Cowork can compile competitive intelligence, build presentations and outline project milestones, creating a coordinated workflow for teams.

Microsoft emphasises that the system operates within enterprise security boundaries. Identity, compliance policies and data permissions remain enforced while tasks execute in a protected cloud environment.

The platform also reflects a multi-model strategy, combining Microsoft AI capabilities with Anthropic technology through the integration of the model behind Claude.

Copilot Cowork is currently available to a limited group of customers through a research preview.

Wider availability is expected later in 2026 through Microsoft’s Frontier programme as the company expands AI-driven workplace automation.

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

Smart Classrooms initiative transforms learning in 10 Thai pilot schools

Ten pilot schools in Buriram and Si Sa Ket provinces have launched Smart Classrooms under the UNESCO–Huawei TEOSA initiative, supporting Thailand’s drive to expand digital education.

Led by UNESCO Bangkok in partnership with Thailand’s Ministry of Education and Huawei Technologies Co., Ltd, the Smart Classrooms initiative aims to strengthen digital learning environments, equip teachers with digital and AI competencies, and support policy development for AI in education. The programme also supports Thailand’s ‘Transforming Education in the Digital Era’ policy and the National AI Strategy and Action Plan (2022–2027).

Each province has one designated ‘mother school’ that serves as a regional digital hub, supporting four surrounding ‘child schools’ by sharing resources, training, and expertise. The ten pilot schools in total have received high-speed internet, interactive digital displays, and collaborative learning platforms that support real-time content sharing and blended learning. Forty-five teachers from the pilot schools also participated in hands-on demonstrations of Smart Classrooms systems on 4–5 March.

‘This new technology will help translate theory into practice, allowing students to experiment, test strategies, and see results immediately,’ said Pathanapong Momprakhon, Principal of Paisan Pittayakom School. UNESCO Bangkok’s Deputy Director and Chief of Education, Marina Patrier, highlighted the importance of combining infrastructure with teacher capacity-building.

‘At UNESCO, we are committed to promoting the ethical and inclusive use of AI in ways that empower teachers and expand opportunities for every learner,’ Ms Patrier said at the launch. ‘While Smart Classrooms provide important tools, it is teachers’ creativity, professional judgement and leadership that ultimately bring these innovations to life.’

Chitralada Chanyaem of the Thai National Commission for UNESCO highlighted the importance of collaboration in advancing digital education.

‘The UNESCO–Huawei Funds-in-Trust Project on Technology-Enabled Open Schools for All stands as a powerful example of collaboration dedicated to transforming education into a system that is open, inclusive, flexible, and resilient in the face of a rapidly changing world, she said. ‘As the future of education cannot be confined within classroom walls, it must bridge sectors and communities, working collaboratively to create equitable and sustainable opportunities for all.’

Teachers observed Huawei technical staff and master teachers demonstrate how digital tools and AI-supported applications can be used in everyday lessons. Ms Piyaporn Kidsirianan, Public Relations Manager at Huawei Technologies (Thailand) Co., Ltd, said the initiative aims to reduce digital inequality.

‘The Open Schools for All initiative represents a commitment to using technology as a bridge to deliver quality education to remote and underserved communities.’ The TEOSA Smart Classrooms initiative combines policy support, digital infrastructure upgrades, and teacher training to help translate Thailand’s digital education ambitions into practical impact at the school level.

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

Space startup to test crypto mining in orbit

Starcloud, a space startup, is preparing to test Bitcoin mining in orbit with its upcoming Starcloud-2 satellite. The mission will carry specialised ASIC mining processors, marking one of the first attempts to run crypto infrastructure beyond Earth.

The initiative builds on a successful 2025 demonstration when Starcloud operated Nvidia H100 GPUs in low Earth orbit. During that mission, the satellite performed AI computing tasks, proving that data-centre-grade hardware can function in space.

Starcloud-2 will expand these capabilities by adding a larger GPU cluster and mining-specific ASICs.

Operating in orbit offers potential advantages for energy-intensive computing. Satellite solar arrays provide near-continuous power, and space’s vacuum allows natural heat dissipation, cutting the need for water-based cooling systems.

Engineers warn that technical challenges remain. Radiation exposure, shielding needs, and the difficulty of repairing hardware once launched could complicate operations.

Despite these obstacles, Starcloud sees orbit as a promising environment for next-generation computing and Bitcoin mining.

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