Researchers found that some large language models can analyse language like a human linguistics graduate. The models diagram sentences, resolve ambiguities and process recursive structures, showing advanced metalinguistic abilities.
The study used specially crafted sentences and invented mini-languages to prevent memorisation. OpenAI’s o1 model correctly applied complex syntactic and phonological rules for entirely new languages.
Experts say the results challenge long-held assumptions about human uniqueness in language. The models have yet to produce original insights, but their reasoning skills match graduate-level performance.
Findings suggest AI may eventually surpass humans in linguistic analysis. Researchers believe continued progress will enable models to generalise better, learn from less data, and handle language creatively.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Vietnam’s 5G network now reaches approximately 59 percent of the population, slightly over one year after commercial services launched in October 2024.
Government data presented at Internet Day 2025 show that Vietnam ranks 10th globally for fixed broadband speed and 15th for mobile broadband, reflecting rapid improvements in national connectivity.
Officials described the Internet as a second living space for citizens, with nearly 80 million users spending an average of seven hours online each day for work, education and social interaction.
Authorities highlighted that expanded 5G coverage supports the development of a digital economy, e-government services and a more connected digital society.
Alongside infrastructure growth, policymakers stressed the need for stronger digital trust.
Vietnam is shifting towards clearer legal frameworks instead of reliance on voluntary self-regulation, while prioritising cybersecurity, data governance and protection against online fraud, deepfakes and AI-driven deception to sustain long-term digital transformation.
Would you like to learn more aboutAI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Leading AI researcher Yann LeCun has argued that large language models only simulate understanding rather than genuinely comprehending the world. Their intelligence, he said, lacks grounding in physical reality and everyday common sense.
Despite being trained on vast amounts of online text, LLMs struggle with unfamiliar situations, according to LeCun. Real-world experience, he noted, provides richer learning than language alone ever could.
Drawing on decades in AI research, LeCun warned that enthusiasm around LLMs mirrors earlier hype cycles that promised human-level intelligence. Similar claims have repeatedly failed to deliver since the 1950s.
Instead of further scaling language models, LeCun urged greater investment in ‘world models’ that can reason about actions and consequences. He also cautioned that current funding patterns risk sidelining alternative approaches to AI.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
AI is entering a new phase, with 2026 expected to mark a shift from experimentation to real-world collaboration. Microsoft executives describe AI as an emerging partner that amplifies human expertise rather than replacing it.
Microsoft says the impact is becoming visible across healthcare, software development, and scientific research. AI tools embedded in Microsoft products are supporting diagnosis, coding, and research workflows.
With the expansion of AI agents across all platforms, organisations are strengthening safeguards to manage new risks. Security leaders argue agents will require clear identities, restricted access, and continuous monitoring.
Microsoft also points to changes in the infrastructure powering AI. The company says future systems will prioritise efficiency and intelligence output, supported by distributed and hybrid cloud architectures.
Looking further ahead, the convergence of AI, supercomputing, and quantum technologies stands out as the main highlight. Hybrid approaches, the company says, are bringing practical quantum advantage closer for applications in materials science, medicine, and research.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Africa cannot realistically compete with the massive capital and computing resources driving frontier AI research in the United States and China, and it does not need to do so.
Instead, Nicholas Okumu contends, the continent’s AI strategy should pivot toward building efficient, practical systems tailored to local needs, from healthcare triage tools in referral hospitals to agriculture, education and public finance solutions grounded in African contexts.
Large, resource-intensive models require infrastructure and ecosystems that most African nations cannot marshal, but smaller, efficient models can perform high-value, domain-specific tasks on ordinary hardware.
Drawing on events from innovation forums and real-world examples, the columnist argues that Africa’s historical experience of innovation under constraint positions it well to lead in relevant, efficient AI applications rather than replicating the ambitions of frontier labs.
The article outlines a three-phase pathway: first, building foundational datasets governed by national or regional frameworks; second, deploying AI where it can deliver transformative value; and third, scaling successful tools to regions with similar development constraints.
If this strategy is followed, the piece argues, African-designed AI systems, particularly those that work well in low-resource environments, could become globally valuable.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Cybersecurity researchers are urging greater caution as Christmas approaches, warning that seasonal scams are multiplying rapidly. Check Point has recorded over 33,500 festive phishing emails and more than 10,000 deceptive social ads within two weeks.
AI tools are helping criminals craft convincing messages that mirror trusted brands and local languages. Attackers are also deploying fake e-commerce sites with AI chatbots, as well as deepfake audio and scripted calls to strengthen vishing attempts.
Smishing alerts imitating delivery firms are becoming more widespread, with recent months showing a marked rise in fraudulent parcel scams. Victims are often tricked into sharing payment details through links that imitate genuine logistics updates.
Experts say fake shops and giveaway scams remain persistent risks, frequently launched from accounts created within the past three months. Users are being advised to ignore unsolicited links, verify retailers and treat unexpected offers with scepticism.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Amazon has published the most frequently asked questions to its Alexa virtual assistant during 2025, providing insight into how people interact with voice-activated AI throughout the year.
Practical questions, such as cooking tips like ‘How long do I poach an egg for?’ and basic science queries like the diameter of Earth, topped the list, showing that many users rely on Alexa for everyday information.
The report also revealed regional and topical variety: in Australia, users asked about sleep help and food classification (e.g. whether a tomato is a fruit), while global queries included questions about celebrities’ heights, weights and net worth.
One of the year’s most frequently asked questions was ‘What does AI stand for?’, indicating ongoing curiosity about the technology that powers the assistant itself. Music and entertainment featured prominently, with Taylor Swift identified as the most played artist of the year and the song ‘APT’ cited as the most played track in multiple regions.
These usage patterns reflect how voice assistants have become integrated into daily routines, from practical tasks to leisure and curiosity-driven searches.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
The US tech company, Oracle, has expanded Oracle Database@Google Cloud to India, making the service available through Google Cloud’s Mumbai region.
Enterprises can access Oracle Exadata, Autonomous AI Database and AI Lakehouse services while keeping data in the region to meet sovereignty and regulatory requirements.
The multicloud offering allows organisations to combine Oracle enterprise data with Google Cloud analytics and AI tools, including BigQuery, Vertex AI and Gemini models.
Customers can modernise applications and migrate mission-critical workloads without sacrificing performance, security or low-latency access.
Oracle Database@Google Cloud is available through the Google Cloud Marketplace, enabling customers to procure services via trusted partners instead of navigating complex contracting models.
Oracle and Google Cloud partners can also integrate the service into broader multicloud solutions.
The launch reflects growing demand for flexible multicloud architectures in India, supporting AI-driven innovation, advanced analytics and accelerated IT modernisation across regulated and data-intensive industries.
Would you like to learn more aboutAI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Scientists at the Icahn School of Medicine at Mount Sinai have created an AI tool called Variant to Phenotype (V2P) that can identify genetic mutations and predict the diseases they may cause, bolstering the field of genetic diagnostics.
The V2P method is designed to accelerate diagnosis and facilitate the discovery of new treatments for complex and rare diseases by comprehensively interpreting genomic data, surpassing the limitations of traditional techniques that often focus solely on mutation detection without predicting phenotypic effects.
This innovation could enhance clinical decision-making by linking specific genetic variants directly to disease risk, helping clinicians prioritise variants for further study and informing patients about likely outcomes sooner.
The findings were published online in Nature Communications, marking a notable advancement in how AI can support precision medicine and research for rare diseases.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Libraries Connected, supported by a £310,400 grant from the UK Government’s Digital Inclusion Innovation Fund administered by the Department for Science, Industry and Technology (DSIT), is launching Innovating in Trusted Spaces: Libraries Advancing the Digital Inclusion Action Plan.
The programme will run from November 2025 to March 2026 across 121 library branches in Newcastle, Northumberland, Nottingham City and Nottinghamshire, targeting older people, low-income families and individuals with disabilities to ensure they are not left behind amid rapid digital and AI-driven change.
Public libraries are already a leading provider of free internet access and basic digital skills support, offering tens of thousands of public computers and learning opportunities each year. However, only around 27 percent of UK adults currently feel confident in recognising AI-generated content online, underscoring the need for improved digital and media literacy.
The project will create and test a new digital inclusion guide for library staff, focusing on the benefits and risks of AI tools, misinformation and emerging technologies, as well as building a national network of practice for sharing insights.
Partners in the programme include Good Things Foundation and WSA Community, which will help co-design materials and evaluate the initiative’s impact to inform future digital inclusion efforts across communities.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
