Digital identities Archives | Digital Watch Observatory

Brazil study maps age assurance practices across 25 digital services

A new study by CGI.br and NIC.br examines how digital services in Brazil implement age assurance measures. Presented in Brasília during an event on the Digital Child and Adolescent Statute (ECA Digital), the study reviewed 25 popular online services used by children and adolescents.

The study found that most of the services analysed do not apply age checks at the point of registration, including some platforms aimed at adults. According to the release, age assurance usually appears later, when users try to access specific features such as livestreaming or monetisation.

Titled ‘Age assurance practices in 25 digital services used by children in Brazil’, the study analysed governance documents published before the ECA Digital entered into force. From 18 March, the law requires information-society services aimed at children and adolescents in Brazil, or likely to be accessed by them, to adopt effective age-assurance measures and parental supervision.

The study found that 11 of the 25 platforms relied on third-party age-assurance services, particularly social media and generative AI platforms. Official identity document submission was the most common verification method, while selfie-based checks were the most common age-estimation tool. Differences were also found between the minimum ages stated by services and those listed in app stores, and some adult-oriented platforms could still be accessed by younger users with parental consent.

Parental supervision tools were available in 15 of the 25 services, but activation was usually optional and depended on parents or guardians. Transparency also emerged as a weakness: only six services published Brazil-specific reports, and only one explained how its minimum-age policy was applied. Policies were often spread across multiple pages, averaging 22 pages per service, and around 40% of the services provided related information in other languages.

Fábio Senne, General Research Coordinator at Cetic.br | NIC.br, said: ‘One of the study’s central aims was to verify the integrity of the information made available by digital services in Brazil. It is essential that data on age protection be communicated clearly and accessibly, allowing more informed and effective parental supervision.’

Juliana Cunha, manager of the Digital Public Policy Advisory Office at CGI.br | NIC.br, said: ‘This survey was developed to support the debate on implementation of the ECA Digital and to offer a clear understanding of the current landscape. This initiative forms part of a broader set of actions by CGI.br and NIC.br aimed at providing technical evidence to support effective enforcement of the law. Our commitment is to foster a safer and more responsible digital ecosystem for children and adolescents in Brazil.’

The release says the study used as a methodological reference the OECD technical paper ‘Age assurance practices of 50 online services used by children’, published in 2025. Information was collected between 10 and 30 January 2026 from public documents made available by the services in Brazil, totalling 550 pages analysed. The event also marked the launch of TIC Kids Online Brazil 2025, a publication on internet use by children and adolescents aged 9 to 17 in Brazil.

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

South Korea sets ambition to become AI leader

South Korea has unveiled a national strategy to become one of the world’s top three AI powers by 2028. The plan combines investment in digital infrastructure, data systems and next-generation connectivity.

Authorities aim to expand networks by advancing 5G capabilities and preparing for the commercial deployment of 6G by 2030. Cybersecurity and data integration are also key priorities to support a stronger digital ecosystem.

The strategy includes developing talent across education levels and investing in core technologies such as semiconductors and quantum computing. AI adoption is expected to expand across sectors, including manufacturing, healthcare and agriculture.

The South Korean officials also plan to promote digital inclusion through learning centres and assistive technologies. Coordination between ministries will be strengthened to ensure effective delivery of the long-term roadmap.

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

Boston schools expand AI learning initiative

A new partnership led by the City of Boston aims to expand AI literacy across public schools, supported by funding from tech entrepreneur Paul English. The initiative brings together government, academia and industry to strengthen digital skills.

The programme will introduce AI-focused learning in high schools, alongside teacher training and the development of industry-informed curricula. Plans include creating student ambassador roles and offering access to advanced courses.

University of Massachusetts Boston in the US will help design educational content and provide resources through its applied AI institute. The collaboration aims to prepare students for changing job markets shaped by emerging technologies.

Officials say the effort will support responsible and ethical use of AI while opening career pathways. An advisory board of industry experts will guide the programme and connect schools with the wider technology sector.

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

Campaign highlights risks of profit-driven digital platforms

A global campaign led by the Norwegian Consumer Council (NCC) has drawn attention to the decline in quality across digital platforms, a phenomenon widely referred to as ‘enshitification’, in which services deteriorate over time as companies prioritise monetisation over user experience.

The initiative has gained momentum through a viral video and coordinated advocacy efforts across multiple regions.

Inshitification is a term coined by journalist Cory Doctorow that describes a pattern in which platforms initially serve users well, then shift towards extracting value from both users and business partners.

In practice, it often results in increased advertising, paywalls, and reduced functionality, with platforms leveraging user dependence to introduce less favourable conditions.

More than 70 advocacy groups across the EU, the US and Norway have urged policymakers to take stronger action, arguing that declining competition and market concentration allow platforms to degrade services without losing users.

Network effects and high switching costs further limit consumer choice, making it difficult to move to alternative platforms even when dissatisfaction grows.

Existing frameworks, such as the Digital Markets Act and the Digital Services Act, aim to address some of these issues by promoting interoperability, transparency, and accountability.

However, experts argue that enforcement remains too slow and insufficient to deter harmful practices, suggesting that stronger regulatory intervention will be necessary to restore balance between consumers, platforms, and competition in the digital economy.

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

Stanford study warns about the risks of ‘sycophantic’ AI chatbots

A new study from Stanford University has raised concerns about the growing use of AI chatbots for personal advice, highlighting risks linked to a behaviour known as ‘sycophancy’, where systems validate users’ views instead of challenging them.

Researchers argue that such responses are not merely stylistic but have broader consequences for decision-making and social behaviour.

The analysis examined multiple leading models, including ChatGPT, Claude, and Gemini, and found that chatbot responses supported user perspectives far more often than human feedback.

In scenarios involving questionable or harmful actions, systems frequently endorsed behaviour that human evaluators would criticise, raising concerns about reliability in sensitive contexts such as relationships or ethical decisions.

Further experiments involving thousands of participants showed that users tend to prefer and trust sycophantic responses, increasing the likelihood of repeated use.

However, such interactions also appeared to reinforce self-centred thinking and reduce willingness to reconsider or apologise, suggesting a deeper impact on social judgement and interpersonal skills.

Researchers warn that users’ tendency to favour agreeable responses may create incentives for developers to prioritise engagement over accuracy or ethical balance.

The findings highlight the need for oversight and caution, with experts advising against relying on AI systems as substitutes for human guidance in complex personal situations.

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

Major service disruption affects DeepSeek chatbot in China

DeepSeek’s chatbot suffered a seven-hour-plus disruption in China, prompting multiple updates as the company worked to restore full functionality. Users began reporting issues on Sunday evening, with further performance problems recorded on Monday morning.

Initial alerts appeared on monitoring platforms and DeepSeek’s own status page, which acknowledged an incident shortly after it began. Although early fixes were deployed within hours, additional disruptions followed, requiring further corrective updates before the system stabilised.

The company has not disclosed the cause of the outage, and no official comment has been provided. The extended downtime stands out for a platform known for consistent performance, which has maintained a near 99 percent uptime record since the launch of its R1 model in 2025.

The disruption comes at a time of heightened anticipation for DeepSeek’s next major update, as speculation builds across China’s competitive AI sector, where firms continue to race to release new models.

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

EU investigates cyber attack targeting Commission websites

The European Commission has confirmed a cyber-attack targeting its cloud infrastructure hosting the Europa.eu services, with authorities acting swiftly to contain the incident and prevent disruption to public access.

The attack was identified on 24 March, prompting immediate mitigation measures to secure systems and maintain service continuity.

Preliminary findings indicate that some data may have been accessed from affected websites, although the full scope of the incident remains under investigation.

The Commission has begun notifying the relevant EU entities that may be affected, while continuing efforts to assess the extent of the breach and strengthen safeguards.

Officials confirmed that internal systems were not affected, limiting the overall impact of the attack.

Monitoring efforts remain ongoing, with additional security measures being implemented to protect data and infrastructure, rather than relying solely on existing defences. The Commission has also committed to analysing the incident to improve its cybersecurity capabilities.

The attack comes amid growing cyber and hybrid threats targeting European institutions and critical services.

Existing frameworks, including the NIS2 Directive and the Cyber Solidarity Act, aim to strengthen resilience and coordination across member states, supporting a more unified response to large-scale cyber incidents across the EU.

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

AI platform from Fujitsu transforms legacy code analysis

Fujitsu has launched a generative AI service that modernises legacy IT systems by analysing source code and generating design documents. The Application Transform platform, powered by Fujitsu Kozuchi, targets complex environments such as COBOL-based enterprise systems.

The service aims to significantly reduce the time and expertise required for system documentation, cutting workloads by up to 97 percent. Fujitsu combines proprietary code analysis with Knowledge Graph-enhanced retrieval to improve accuracy and reduce missing or inconsistent outputs.

Enhanced by generative AI, the system produces structured, readable documentation while ensuring consistency across large, complex codebases. Reported improvements include higher comprehensiveness and significantly better readability compared with conventional methods.

Fujitsu plans to offer the service as SaaS in Japan from 30 March 2026, with additional capabilities such as automated code rewriting and system maintenance support expected in future updates.

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

VivaCity partners with Nottingham to enhance urban transport using AI

Nottingham City Council has partnered with VivaCity to install over 200 AI-enabled transport sensors across the city. The sensors include ANPR, traffic monitoring, and Smart Signal Control capabilities.

Sensors will collect real-time, anonymous data on vehicle types, pedestrians, and cyclists to inform traffic management decisions. The first Smart Junction at the Ring Road-Aspley Lane will adjust traffic lights according to current conditions.

Funding comes from the Future Transport Zones Fund, for which the Department awarded £16.7 million for Transport. Installation began in February 2023 and will finish by November 2023, with coverage across main routes.

Data from the sensors will feed into a public Data Hub alongside car park and EV charging datasets. Air quality monitors will be added near sensors to help assess correlations between road use and pollution levels.

Sensors will not function as speed cameras and will not record personal information. The technology will be upgraded over time to identify additional vehicle types such as taxis, minibuses, and mobility scooters.

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

UK regulator targets misleading online reviews in new crackdown

The Competition and Markets Authority has launched new investigations into five companies as part of a wider crackdown on fake and misleading online reviews, targeting practices that shape consumer decisions rather than reflect genuine customer experiences.

The cases involve Autotrader, Feefo, Dignity, Just Eat and Pasta Evangelists across sectors, including car sales, food delivery and funeral services.

CMA is examining whether negative reviews were suppressed, ratings inflated, or incentives offered in exchange for positive feedback without disclosure.

Concerns also extend to moderation practices and whether review systems provide a complete and accurate picture of customer experiences, rather than favouring reputational or commercial interests. No conclusions have yet been reached on whether consumer law has been breached.

Online reviews play a central role in consumer behaviour, influencing significant levels of spending across the UK economy.

Research indicates that a large majority of consumers rely on reviews when making purchasing decisions, raising concerns that misleading content can distort markets and undermine trust, particularly as AI makes it harder to detect fabricated reviews.

The investigations form part of a broader enforcement effort under the Digital Markets Competition and Consumers Act 2024, which introduced stricter rules on fake and misleading reviews.

Authorities aim to improve transparency and accountability across digital platforms, with potential penalties reaching up to 10% of global turnover for companies found to have breached consumer protection laws.

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