Digital identities Archives | Page 26 of 180

China expands oversight of youth online safety

China has introduced new measures to regulate online information that could affect the physical and mental health of minors. Authorities in China said the rules will take effect on 1 March and aim to improve protection for young internet users.

The regulators identified four categories of online information that may harm minors. The authorities have also addressed emerging risks linked to algorithmic recommendations and generative AI technologies.

The framework in China requires internet platforms and content creators to prevent and respond to harmful material. Regulators said companies must strengthen the monitoring and governance of content affecting minors.

Authorities said the measures are designed to create a cleaner online environment for children. Officials also stressed greater responsibility for platforms that manage digital content used by minors.

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

OpenAI upgrades ChatGPT conversations with GPT-5.3 Instant

The most widely used ChatGPT model has received an update from OpenAI, introducing GPT-5.3 Instant to make everyday conversations more coherent, useful, and natural.

An upgrade that focuses on improving tone, contextual understanding, and the flow of dialogue rather than only benchmark performance.

One of the main improvements concerns how the model handles refusals and safety responses. Earlier versions sometimes declined questions that could have been answered safely or delivered overly cautious explanations before responding.

GPT-5.3 Instant instead gives more direct answers while still maintaining safety constraints, reducing interruptions that previously slowed conversations.

The update also improves the way ChatGPT uses information from the web. Instead of simply summarising search results or presenting long lists of links, the model now integrates online information with its own reasoning.

Such an approach aims to produce more relevant answers that highlight key insights at the beginning of responses.

Reliability has also improved. Internal evaluations conducted by OpenAI show reductions in hallucination rates across multiple domains.

When using web sources, hallucinations dropped by roughly 26.8 percent in higher-risk fields such as medicine, law, and finance. Improvements were also recorded when the model relied only on its internal knowledge.

Beyond factual accuracy, the model is designed to feel more natural in conversation. OpenAI says the system now avoids overly preachy language, unnecessary disclaimers, and intrusive remarks that previously disrupted dialogue.

The goal is a more consistent conversational personality across updates, while maintaining the familiar user experience of ChatGPT.

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

EU citizens propose public social media network under new initiative

The European Commission has registered a European Citizens’ Initiative proposing the creation of a public social media platform operating at the European level, rather than relying exclusively on private technology companies.

An initiative titled the European Public Social Network calls for legislation establishing a publicly funded digital platform designed to serve societal interests.

Organisers argue that a publicly owned network could function independently from commercial incentives and political pressure while guaranteeing equal rights for users across the EU. The proposed platform would operate as a public service overseen by society rather than private corporations.

Registration confirms that the proposal meets the legal requirements of the European Citizens’ Initiative framework. The Commission has not yet assessed the substance of the idea, and registration does not imply support for the proposal.

Supporters must now gather 1 million signatures from citizens across at least 7 EU member states within 12 months. If the threshold is reached, the Commission will be required to formally examine the initiative and decide whether legislative action is appropriate.

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

Guterres convenes global UN panel of 40 experts to assess AI risks

UN Secretary-General António Guterres told the inaugural meeting of a new independent group of experts on AI convened by the UN that they have a huge responsibility to help shape how the technology is used ‘for the benefit of humanity’.

‘Individually, you come from diverse regions and disciplines, bringing outstanding expertise in AI and related fields. Collectively, you represent something the world has never seen before,’ the UN chief told scientists on Tuesday at the first meeting of the Independent International Scientific Panel on AI.

The panel brings together 40 experts who aim to help close ‘the AI knowledge gap’ and assess the real impact the frontier technology will have across economies and societies so that countries can act with the same ‘clarity’ on a level playing field.

The experts will provide scientific assessments independent of any government, company or institution – including the UN itself. ‘AI is advancing at lightning speed… no country, no company, and no field of research can see the full picture alone,’ Guterres said. ‘The world urgently needs a shared, global understanding of artificial intelligence; grounded not in ideology, but in science.’

Warning about the stakes involved as AI evolves rapidly, Guterres said the technology will shape peace and security, human rights, and sustainable development for decades to come. ‘I have seen how quickly fear can take hold when facts are missing or distorted – how trust breaks down, and division deepens,’ he said. At a time when ‘geopolitical tensions are rising, and conflicts are raging,’ he stressed that the need for shared understanding and ‘safe and responsible AI could not be greater.’

As AI development accelerates, the Secretary-General also warned the panel that it is ‘in a race against time.’ Addressing concerns about the pace of technological change, he said: ‘Never in the future will we move as slowly as we are moving now. We are indeed in a high level of acceleration.’

Guterres also pointed to earlier work through the UN High-Level Advisory Body on AI, noting that the new scientific panel does not ‘start from zero’. Concluding his remarks, Guterres told the experts: ‘I can think of no more important assignment for our world today.’

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

Australia reviews children’s social media ban

Australia has begun reviewing its ban on social media accounts for children under 16, introduced in December 2025. Australia’s eSafety Commissioner is tracking more than 4,000 children and families to assess how the policy works in practice.

Researchers in Australia will analyse surveys, interviews and voluntary smartphone data to measure how young people interact with apps. Officials in Australia aim to understand how the ban affects children, parents and everyday online behaviour.

Early reactions in Australia have been mixed, with some teenagers telling media outlets they bypass age verification systems. Platforms reportedly remain accessible to some minors in Australia.

Meanwhile, the UK government has launched a public consultation on potential social media restrictions for children. Policymakers in the UK are seeking views on bans, stronger age verification and limits on addictive platform features.

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

EU considers placing Roblox under strict Digital Services Act rules

European regulators are examining whether Roblox should fall under the Digital Services Act’s most stringent obligations rather than remain outside the bloc’s most demanding platform rules.

The European Commission began analysing the gaming platform’s reported user figures after the company disclosed roughly 48 million monthly users across the EU.

Numbers above the threshold could qualify Roblox as a Very Large Online Platform under the DSA. Such a designation would mark the first time a gaming platform enters the category alongside social media services already subject to heightened oversight.

Platforms receiving the label must conduct regular risk assessments, submit mitigation reports and demonstrate stronger safeguards for minors.

Regulatory pressure has already begun at the national level. The Dutch Authority for Consumers and Markets launched an investigation in January after concerns that children could encounter violent or sexually explicit content within Roblox games or interact with harmful actors through online features.

Designation at the EU level would transfer supervisory authority to the European Commission, enabling wider investigations and potential fines if violations occur. Officials are still verifying user data before making a formal decision, and no deadline has been announced for the process.

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

OneTrust’s new CEO outlines AI governance ambitions

OneTrust has entered a new leadership phase in the US after appointing John Heyman as chief executive, replacing founder Kabir Barday. Barday will remain on the board in an advisory role as the US-based compliance technology firm continues to push into AI governance.

John Heyman said organisations across the US and globally are rapidly integrating AI into daily operations. Companies deploying large numbers of AI agents increasingly need tools to manage risk, data use and regulatory compliance.

OneTrust believes demand for governance technology will grow as AI systems multiply inside businesses in the US and worldwide. John Heyman described a future where automated monitoring tools oversee AI agents operating within company systems.

Leadership at OneTrust in the US aims to build systems that track how AI agents collect and share data while maintaining enterprise control. Growing adoption of AI in the US and globally continues to drive demand for responsible governance platforms.

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

AI tool from MIT speeds up complex engineering optimisation

MIT researchers have developed a new AI approach that helps engineers solve complex design problems faster, from power grid optimisation to vehicle safety.

The method adapts a foundation model trained on tabular data, enabling high-dimensional optimisation without retraining and significantly speeding up results.

The system uses a foundation model with Bayesian optimisation to pinpoint the variables that most impact outcomes. Focusing on key variables, the model finds top solutions 10 to 100 times faster than existing optimisation methods.

Early tests show the approach excels in costly, time-consuming scenarios like car crash testing and power system design. The technique lowers computational demands and suits large-scale, high-frequency engineering challenges across multiple domains.

Researchers aim to expand the method to even higher-dimensional problems, such as naval ship design, while highlighting the broader potential of foundation models as algorithmic engines in scientific and engineering tools.

Experts see it as a practical step toward making advanced optimisation more accessible in real-world applications.

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

Developers gain early access to Gemini 3.1 Flash-Lite

Google’s Gemini 3.1 Flash-Lite has launched in preview for developers via AI Studio and for enterprises through Vertex AI. Designed for high-volume workloads, it promises fast, cost-effective performance while maintaining high-quality outputs.

Priced at just $0.25 per million input tokens and $1.50 per million output tokens, 3.1 Flash-Lite offers 2.5X faster response times and 45% higher output speed than the previous 2.5 Flash model.

Benchmarks show strong performance across reasoning and multimodal tasks, including an Elo score of 1432 on Arena.ai, 86.9% on GPQA Diamond, and 76.8% on MMMU Pro, surpassing some older, larger Gemini models.

The model also provides adaptive intelligence features, allowing developers to adjust how much the AI ‘thinks’ for each task. The model handles both high-frequency tasks, such as translation, and complex tasks, such as interface generation and simulations.

Early-access developers and companies report that 3.1 Flash-Lite handles complex workloads with precision comparable to larger models. Its speed, affordability, and reasoning capabilities make it an attractive choice for scalable, real-time AI applications.

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

Chrome moves to rapid releases as Google responds to AI disruption

Google is accelerating Chrome’s release cycle rather than maintaining its long-standing four-week cadence.

From September, users on desktop and mobile platforms will receive new stable versions every two weeks, doubling the frequency of feature milestones across speed, stability and usability. Weekly security updates introduced in 2023 remain unchanged.

The faster pace comes as AI-driven browsers seek a foothold in a market long dominated by Chrome.

Products, such as ChatGPT Atlas and Perplexity’s Comet, embed agentic assistants directly into the browsing experience, automating tasks from summarising pages to scheduling meetings.

Chrome has responded with deeper Gemini integration, including the rollout of autonomous features across its interface.

Google maintains that the accelerated schedule reflects the needs of the evolving web platform, arguing that developers require quicker access to updated tools.

Yet the timing aligns with growing competitive pressure from AI-native browsers, prompting speculation that Chrome’s dominance can no longer be taken for granted.

The shift will begin with Chrome version 153 in beta and stable channels on 8 September 2026. Enterprise administrators and Chromebook users will continue to rely on the eight-week Extended Stable branch, which remains unchanged for organisations that need slower, controlled deployments.

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