Capacity development

UNESCO launches AI guidelines for courts and tribunals

UNESCO has launched new Guidelines for the Use of AI Systems in Courts and Tribunals to ensure AI strengthens rather than undermines human-led justice. The initiative arrives as courts worldwide face millions of pending cases and limited resources.

In Argentina, AI-assisted legal tools have increased case processing by nearly 300%, while automated transcription in Egypt is improving court efficiency.

Judicial systems are increasingly encountering AI-generated evidence, AI-assisted sentencing, and automated administrative processes. AI misuse can have serious consequences, as seen in the UK High Court where false AI-generated arguments caused delays, extra costs, and fines.

UNESCO’s Guidelines aim to prevent such risks by emphasising human oversight, auditability, and ethical AI use.

The Guidelines outline 15 principles and provide recommendations for judicial organisations and individual judges throughout the AI lifecycle. They also serve as a benchmark for developing national and regional standards.

UNESCO’s Judges’ Initiative, which has trained over 36,000 judicial operators in 160 countries, played a key role in shaping and peer-reviewing the Guidelines.

The official launch will take place at the Athens Roundtable on AI and the Rule of Law in London on 4 December 2025. UNESCO aims for the standards to ensure responsible AI use, improve court efficiency, and uphold public trust in the judiciary.

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

FCA launches AI Live Testing for UK financial firms

The UK’s Financial Conduct Authority has launched an AI Live Testing initiative to help firms safely deploy AI in financial markets. Major companies, including NatWest, Monzo, Santander, Scottish Widows, Gain Credit, Homeprotect, and Snorkl, are participating in the first cohort.

Firms receive tailored guidance from the FCA and its technical partner, Advai, to develop and assess AI applications responsibly.

AI testing focuses on retail financial services, exploring uses such as debt resolution, financial advice, improving customer engagement, streamlining complaints handling, and supporting smarter spending and saving decisions.

The project aims to answer key questions around evaluation frameworks, governance, live monitoring, and risk management to protect both consumers and markets.

Jessica Rusu, FCA chief data officer, said the initiative helps firms use AI safely while guiding the FCA on its impact in UK financial services. The project complements the FCA’s Supercharged Sandbox, which supports firms in earlier experimentation phases.

Applications for the second AI Live Testing cohort open in January 2026, with participating firms able to start testing in April. Insights from the initiative will inform FCA AI policy, supporting innovation while ensuring responsible deployment.

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

Texas makes historic investment with $5 million Bitcoin purchase

Texas has become the first US state to fund a strategic cryptocurrency reserve, purchasing approximately $5 million in Bitcoin through BlackRock’s iShares Bitcoin Trust ETF.

The move follows Governor Greg Abbott signing Senate Bill 21, allowing the comptroller’s office to create a public crypto reserve. states, such as New Hampshire and Arizona, have passed similar bills, but Texas is the first to execute an actual purchase.

The ETF acquisition acts as a temporary measure while the state finalises a contract with a cryptocurrency custodian. Comptroller representatives called the purchase a ‘placeholder investment’ while reviewing bids for a permanent custodian.

Lawmakers have allocated $10 million to the reserve, a small portion of Texas’ $338 billion budget, yet supporters argue it marks an important step for the growing crypto industry.

Bitcoin prices have fluctuated significantly this year, peaking above $126,000 in October before dropping to around $85,000 recently. The state’s purchase at roughly $87,000 per bitcoin reflects ongoing market volatility.

Advocates see the investment as forward-looking, citing potential long-term benefits in job creation, tax revenue, and digital asset adoption.

Critics remain sceptical, warning that public crypto investments carry high risk and may favour industry interests over taxpayers. Some economists criticised the move as conflicting with Texas’ conservative fiscal approach and risky government speculation.

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

Serbia sees wider coverage as Yettel activates 5G

Yettel has launched its 5G network in Serbia, offering higher speeds, lower latency, and support for large numbers of connected devices. Customers need a 5G-ready handset and coverage access, which currently spans major cities and tourist areas. The operator plans wider expansion as deployment progresses.

The service uses recently acquired spectrum, with 5G delivered across the 700 MHz low band and the 3.5 GHz mid band. The frequencies support stronger indoor reach and higher-capacity performance. Yettel says the combination will improve everyday mobile connectivity and enable new digital services.

Use cases include faster downloads, smoother streaming, and more responsive cloud-based gaming. Lower latency will also support remote work and IoT applications. The company expects the network to underpin emerging services that rely on real-time communication and consistent mobile performance.

Yettel forms part of the e& PPF Telecom Group and operates more than 130 retail locations alongside its digital channels. The company says the 5G rollout complements ongoing efforts to modernise national infrastructure. It also aims to maintain strong service quality across urban and regional areas.

The network received the umlaut ‘Best in Test’ award in 2025, marking a ninth consecutive national win. Yettel frames 5G as the next stage of its technological development. The operator expects the upgrade to strengthen the broader digital ecosystem of Serbia and improve long-term connectivity options.

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

New findings reveal untrained AI can mirror human brain responses

Researchers at Johns Hopkins report that brain-inspired AI architectures can display human-like neural activity before any training. Structural design may provide stronger starting points than data-heavy methods. The findings challenge long-held views about how machine intelligence forms.

Researchers tested modified transformers, fully connected networks, and convolutional networks across multiple variants. They compared untrained model responses with neural data from humans and primates viewing identical images. The approach allowed a direct measure of architectural influence.

Transformers and fully connected networks showed limited change when scaled. Convolutional models, by contrast, produced patterns that aligned more closely with human brain activity. Architecture appears to be a decisive factor early in development.

Untrained convolutional models matched aspects of systems trained on millions of images. The results suggest brain-like structures could cut reliance on vast datasets and energy-intensive computation. The implications may reshape how advanced models are engineered.

Further research will examine simple, biologically inspired learning rules. The team plans to integrate these mechanisms into future AI frameworks. The goal is to combine architecture and biology to accelerate meaningful advances.

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

Governments urged to build learning systems for the AI era

Governments are facing increased pressure to govern AI effectively, prompting calls for continuous institutional learning. Researchers argue that the public sector must develop adaptive capacity to keep pace with rapid technological change.

Past digital reforms often stalled because administrations focused on minor upgrades rather than redesigning core services. Slow adaptation now carries greater risks, as AI transforms decisions, systems and expectations across government.

Experts emphasise the need for a learning infrastructure that facilitates to reliable flow of knowledge across institutions. Singapore and the UAE have already invested heavily in large-scale capability-building programmes.

Public servants require stronger technical and institutional literacy, supported through ongoing training and open collaboration with research communities. Advocates say that states that embed learning deeply will govern AI more effectively and maintain public trust.

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

Mistral AI unveils new open models with broader capabilities

Yesterday, Mistral AI introduced Mistral 3 as a new generation of open multimodal and multilingual models that aim to support developers and enterprises through broader access and improved efficiency.

The company presented both small dense models and a new mixture-of-experts system called Mistral Large 3, offering open-weight releases to encourage wider adoption across different sectors.

Developers are encouraged to build on models in compressed formats that reduce deployment costs, rather than relying on heavier, closed solutions.

The organisation highlighted that Large 3 was trained with extensive resources on NVIDIA hardware to improve performance in multilingual communication, image understanding and general instruction tasks.

Mistral AI underlined its cooperation with NVIDIA, Red Hat and vLLM to deliver faster inference and easier deployment, providing optimised support for data centres along with options suited for edge computing.

A partnership that introduced lower-precision execution and improved kernels to increase throughput for frontier-scale workloads.

Attention was also given to the Ministral 3 series, which includes models designed for local or edge settings in three sizes. Each version supports image understanding and multilingual tasks, with instruction and reasoning variants that aim to strike a balance between accuracy and cost efficiency.

Moreover, the company stated that these models produce fewer tokens in real-world use cases, rather than generating unnecessarily long outputs, a choice that aims to reduce operational burdens for enterprises.

Mistral AI continued by noting that all releases will be available through major platforms and cloud partners, offering both standard and custom training services. Organisations that require specialised performance are invited to adapt the models to domain-specific needs under the Apache 2.0 licence.

The company emphasised a long-term commitment to open development and encouraged developers to explore and customise the models to support new applications across different industries.

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

Public backlash grows as Coupang faces scrutiny over massive data leak

South Korea is facing broader concerns about data governance following Coupang’s confirmation of a breach affecting 33.7 million accounts. Investigators say the leak began months before it was detected, highlighting weak access controls and delayed monitoring across major firms.

Authorities believe a former employee exploited long-valid server tokens and unrevoked permissions to extract customer records. Officials say the scale of the incident underscores persistent gaps in offboarding processes and basic internal safeguards.

Regulators have launched parallel inquiries to assess compliance violations and examine whether structural weaknesses extend beyond a single company. Recent leaks at telecom and financial institutions have raised similar questions about systemic risk.

Public reaction has been intense, with online groups coordinating class-action filings and documenting spikes in spam after the exposure. Many argue that repeated incidents show a more profound corporate reluctance to invest meaningfully in security.

Lawmakers are now signalling plans for more substantial penalties and tighter oversight. Analysts warn that unless companies elevate data protection standards, South Korea will continue to face cascading breaches that damage public trust.

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

AI helps detect congenital heart defects in unborn babies

Mount Sinai doctors in New York City are the first to utilise AI to enhance prenatal ultrasounds and detect congenital heart defects more effectively. BrightHeart’s FDA-approved technology is now used at Mount Sinai-affiliated Carnegie Imaging for Women across three Manhattan locations.

Congenital heart defects affect about 1 in 500 newborns and often require urgent intervention.

A study in Obstetrics & Gynecology found AI-assisted ultrasounds detected major defects with over 97 percent accuracy, cut reading time by 18 percent, and raised confidence scores by 19 percent.

The study reviewed 200 fetal ultrasounds from 11 centres across two countries, with and without AI assistance, by obstetricians and maternal-fetal medicine specialists.

AI improved detection, confidence, and efficiency, especially in centres without specialised fetal heart experts.

Experts say AI can level the field of prenatal diagnosis and optimise patient care. Dr Lam-Rachlin and Dr Rebarber emphasised AI’s potential to standardise detection and urged further research for routine clinical use.

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

NVIDIA platform lifts leading MoE models

Frontier developers are adopting a mixture-of-experts architecture as the foundation for their most advanced open-source models. Designers now rely on specialised experts that activate only when needed instead of forcing every parameter to work on each token.

Major models, such as DeepSeek-R1, Kimi K2 Thinking, and Mistral Large 3, rise to the top of the Artificial Analysis leaderboard by utilising this pattern to combine greater capability with lower computational strain.

Scaling the architecture has always been the main obstacle. Expert parallelism requires high-speed memory access and near-instant communication between multiple GPUs, yet traditional systems often create bottlenecks that slow down training and inference.

NVIDIA has shifted toward extreme hardware and software codesign to remove those constraints.

The GB200 NVL72 rack-scale system links seventy-two Blackwell GPUs via fast shared memory and a dense NVLink fabric, enabling experts to exchange information rapidly, rather than relying on slower network layers.

Model developers report significant improvements once they deploy MoE designs on NVL72. Performance leaps of up to ten times have been recorded for frontier systems, improving latency, energy efficiency and the overall cost of running large-scale inference.

Cloud providers integrate the platform to support customers in building agentic workflows and multimodal systems that route tasks between specialised components, rather than duplicating full models for each purpose.

Industry adoption signals a shift toward a future where efficiency and intelligence evolve together. MoE has become the preferred architecture for state-of-the-art reasoning, and NVL72 offers a practical route for enterprises seeking predictable performance gains.

NVIDIA positions its roadmap, including the forthcoming Vera Rubin architecture, as the next step in expanding the scale and capability of frontier AI.

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