Human rights principles Archives | Page 11 of 128

AI features disabled on MEP tablets amid European Parliament security concerns

The European Parliament has disabled AI features on the tablets it provides to lawmakers, citing cybersecurity and data protection concerns. Built-in AI tools like writing and virtual assistants have been disabled, while third-party apps remain mostly unaffected.

The decision follows an assessment highlighting that some AI features send data to cloud services rather than processing it locally.

Lawmakers have been advised to take similar precautions on their personal devices. Guidance includes reviewing AI settings, disabling unnecessary features, and limiting app permissions to reduce exposure of work emails and documents.

Officials stressed that these measures are intended to prevent sensitive data from being inadvertently shared with service providers.

The move comes amid broader European scrutiny of reliance on overseas digital platforms, particularly US-based services. Concerns over data sovereignty and laws like the US Cloud Act have amplified fears that personal and sensitive information could be accessed by foreign authorities.

AI tools, which require extensive access to user data, have become a key focus in ongoing debates over digital security in the EU.

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

Google outlines progress in responsible AI development

Google published its latest Responsible AI Progress Report, showing how AI Principles guide research, product development, and business decisions. Rising model capabilities and adoption have moved the focus from experimentation to real-world industry integration.

Governance and risk management form a central theme of the report, with Google describing a multilayered oversight structure spanning the entire AI lifecycle.

Advanced testing methods, including automated adversarial evaluations and expert review, are used to identify and mitigate potential harms as systems become more personalised and multimodal.

Broader access and societal impact remain key priorities. AI tools are increasingly used in science, healthcare, and environmental forecasting, highlighting their growing role in tackling global challenges.

Collaboration with governments, academia, and civil society is presented as essential for maintaining trust and setting industry standards. Sharing research and tools continues to support responsible AI innovation and broaden its benefits.

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

Shein faces formal proceedings under EU Digital Services Act

The European Commission has opened formal proceedings against Shein under the Digital Services Act over addictive design and illegal product risks. The move follows preliminary reviews of company reports and responses to information requests. Officials said the decision does not prejudge the outcome.

Investigators will review safeguards to prevent illegal products being sold in the European Union, including items that could amount to child sexual abuse material, such as child-like sex dolls. Authorities will also assess how the platform detects and removes unlawful goods offered by third-party sellers.

The Commission will examine risks linked to platform design, including engagement-based rewards that may encourage excessive use. Officials will assess whether adequate measures are in place to limit potential harm to users’ well-being and ensure effective consumer protection online.

Transparency obligations under the DSA are another focal point. Platforms must clearly disclose the main parameters of their recommender systems and provide at least one easily accessible option that is not based on profiling. The Commission will assess whether Shein meets these requirements.

Coimisiún na Meán, the Digital Services Coordinator of Ireland, will assist the investigation as Ireland is Shein’s EU base. The Commission may seek more information or adopt interim measures if needed. Proceedings run alongside consumer protection action and product safety enforcement.

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

EDPS urges stronger safeguards in EU temporary chat-scanning rules

Concerns over privacy safeguards have resurfaced as the European Data Protection Supervisor urges legislators to limit indiscriminate chat-scanning in the upcoming extension of temporary EU rules.

The supervisor warns that the current framework risks enabling broad surveillance instead of focusing on targeted action against criminal content.

The EU institutions are considering a short-term renewal of the interim regime governing the detection of online material linked to child protection.

Privacy officials argue that such measures need clearer boundaries and stronger oversight to ensure that automated scanning tools do not intrude on the communications of ordinary users.

EDPS is also pressing lawmakers to introduce explicit safeguards before any renewal is approved. These include tighter definitions of scanning methods, independent verification, and mechanisms that prevent the processing of unrelated personal data.

According to the supervisor, temporary legislation must not create long-term precedents that weaken confidentiality across messaging services.

The debate comes as the EU continues discussions on a wider regulatory package covering child-protection technologies, encryption and platform responsibilities.

Privacy authorities maintain that targeted tools can be more practical than blanket scanning, which they consider a disproportionate response.

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

China boosts AI leadership with major model launches ahead of Lunar New Year

Leading Chinese AI developers have unveiled a series of advanced models ahead of the Lunar New Year, strengthening the country’s position in the global AI sector.

Major firms such as Alibaba, ByteDance, and Zhipu AI introduced new systems designed to support more sophisticated agents, faster workflows and broader multimedia understanding.

Industry observers also expect an imminent release from DeepSeek, whose previous model disrupted global markets last year.

Alibaba’s Qwen 3.5 model provides improved multilingual support across text, images and video while enabling rapid AI agent deployment instead of slower generation pipelines.

ByteDance followed up with updates to its Doubao chatbot and the second version of its image-to-video tool, SeeDance, which has drawn copyright concerns from the Motion Picture Association due to the ease with which users can recreate protected material.

Zhipu AI expanded the landscape further with GLM-5, an open-source model built for long-context reasoning, coding tasks, and multi-step planning. The company highlighted the model’s reliance on Huawei hardware as part of China’s efforts to strengthen domestic semiconductor resilience.

Meanwhile, excitement continues to build for DeepSeek’s fourth-generation system, expected to follow the widespread adoption and market turbulence associated with its V3 model.

Authorities across parts of Europe have restricted the use of DeepSeek models in public institutions because of data security and cybersecurity concerns.

Even so, the rapid pace of development in China suggests intensifying competition in the design of agent-focused systems capable of managing complex digital tasks without constant human oversight.

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

Meta explores AI system for digital afterlife

Meta has been granted a patent describing an AI system that could simulate a person’s social media activity, even after their death. The patent, originally filed in 2023 and approved in late December, outlines how AI could replicate a user’s online presence by drawing on their past posts, messages and interactions.

According to the filing, a large language model could analyse a person’s digital history, including comments, chats, voice messages and reactions, to generate new content that mirrors their tone and behaviour. The system could respond to other users, publish updates and continue conversations in a way that resembles the original account holder.

The patent suggests the technology could be used when someone is temporarily absent from a platform, but it also explicitly addresses the possibility of continuing activity after a user’s death. It notes that such a scenario would carry more permanent implications, as the person would not be able to return and reclaim control of the account.

More advanced versions of the concept could potentially simulate voice or even video interactions, effectively creating a digital persona capable of engaging with others in real time. The idea aligns with previous comments by Meta CEO Mark Zuckerberg, who has said AI could one day help people interact with digital representations of loved ones, provided consent mechanisms are in place.

Meta has stressed that the patent does not signal an imminent product launch, describing it as a protective filing for a concept that may never be developed. Still, similar services offered by startups have already sparked ethical debate, raising questions about digital identity, consent and the emotional impact of recreating the online presence of someone who has died.

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

AI cheating allegation sparks discrimination lawsuit

A University of Michigan student has filed a federal lawsuit accusing the university of disability discrimination after professors allegedly claimed she used AI to write her essays. The student, identified in court documents as ‘Jane Doe,’ denies using AI and argues that symptoms linked to her medical conditions were wrongly interpreted as signs of cheating.

According to the complaint, Doe has obsessive-compulsive disorder and generalised anxiety disorder. Her lawyers argue that traits associated with those conditions, including a formal tone, structured writing, and consistent style, were cited by instructors as evidence that her work was AI-generated. They say she provided proof and medical documentation supporting her case but was still subjected to disciplinary action and prevented from graduating.

The lawsuit alleges that the university failed to provide appropriate disability-related accommodations during the academic integrity process. It also claims that the same professor who raised the concerns remained responsible for grading and overseeing remedial work, despite what the complaint describes as subjective judgments and questionable AI-detection methods.

The case highlights broader tensions on campuses as educators grapple with the rapid rise of generative AI tools. Professors across the United States report growing difficulty distinguishing between student work and machine-generated text, while students have increasingly challenged accusations they say rely on unreliable detection software.

Similar legal disputes have emerged elsewhere, with students and families filing lawsuits after being accused of submitting AI-written assignments. Research has suggested that some AI-detection systems can produce inaccurate results, raising concerns about fairness and due process in academic settings.

The University of Michigan has been asked to comment on the lawsuit, which is likely to intensify debate over how institutions balance academic integrity, disability rights, and the limits of emerging AI detection technologies.

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

Ireland’s DPC opens data privacy probe into X’s Grok

Ireland’s Data Protection Commission (DPC) has opened a formal investigation into X, focusing on whether the platform complied with its EU privacy obligations after users reportedly generated and shared sexualised, AI-altered images using Grok, the chatbot integrated into X. The inquiry will examine how the EU users’ personal data was processed in connection with this feature, under Ireland’s Data Protection Act and the GDPR framework.

The controversy centres on prompts that can ‘edit’ real people’s photos, sometimes producing non-consensual sexualised imagery, with allegations that some outputs involve children. The DPC has said it has been engaging with X since the reports first emerged and has now launched what it describes as a large-scale inquiry into the platform’s compliance with core GDPR duties.

Public and political reaction has intensified as examples of users altering images posted by others without consent, including ‘undressing’ edits, circulated. Child-safety concerns have widened the issue beyond platform moderation into questions of legality, safeguards, and accountability for generative tools embedded in mass-use social networks.

X has said it has introduced restrictions and safety measures around Grok’s image features, but regulators appear unconvinced that guardrails are sufficient when tools can be repurposed for non-consensual sexual content at scale. The DPC’s inquiry will test, in practical terms, whether a platform can roll out powerful image-generation/editing functions while still meeting the EU privacy requirements for lawful processing, risk management, and protection of individuals.

Why does it matter?

The DPC (Data Protection Commission) is Ireland’s national data protection authority, an Irish public regulator, but at the same time, it operates within the EU’s GDPR system as part of the network of EU/EEA regulators (the ‘supervisory authorities’). The DPC’s probe lands on top of a separate European Commission investigation launched in January under the EU’s Digital Services Act, after concerns that Grok-fuelled deepfakes on X included manipulated sexually explicit images that ‘may amount to child sexual abuse material,’ and questions about whether X properly assessed and mitigated those risks before deployment. Together, the two tracks show how the EU is using both privacy law (GDPR) and platform safety rules (DSA) to pressure large platforms to prove that ‘generative’ features are not being shipped faster than the safeguards needed to prevent serious harm, especially when women and children are the most likely targets.

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

Parliament halts built-in AI tools on tablets and other devices over data risks

The European Parliament has disabled built-in AI features on tablets issued to lawmakers, citing cybersecurity and data protection risks. An internal email states that writing assistants, summarisation tools, and enhanced virtual assistants were turned off after security assessments.

Officials said some AI functions on tablets rely on cloud processing for tasks that could be handled locally, potentially transmitting data off the device. A review is underway to clarify how much information may be shared with service providers.

Only pre-installed AI tools were affected, while third-party apps remain available. Lawmakers were advised to review AI settings on personal devices, limit app permissions, and avoid exposing work emails or documents to AI systems.

The step reflects wider European concerns about digital sovereignty and reliance on overseas technology providers. US legislation, such as the Cloud Act, allows authorities to access data held by American companies, raising cross-border data protection questions.

Debate over AI security is intensifying as institutions weigh innovation against the risks of remote processing and granular data access. Parliament’s move signals growing caution around handling sensitive information in cloud-based AI environments.

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

Study says China AI governance not purely state-driven

New research challenges the view that China’s AI controls are solely the product of authoritarian rule, arguing instead that governance emerges from interaction between the state, private sector and society.

A study by Xuechen Chen of Northeastern University London and Lu Xu of Lancaster University argues that China’s AI governance is not purely top-down. Published in the Computer Law & Security Review, it says safeguards are shaped by regulators, companies and social actors, not only the central government.

Chen calls claims that Beijing’s AI oversight is entirely state-driven a ‘stereotypical narrative’. Although the Cyberspace Administration of China leads regulation, firms such as ByteDance and DeepSeek help shape guardrails through self-regulation and commercial strategy.

China was the first country to introduce rules specific to generative AI. Systems must avoid unlawful or vulgar content, and updated legislation strengthens minor protection, limiting children’s online activity and requiring child-friendly device modes.

Market incentives also reinforce compliance. As Chinese AI firms expand globally, consumer expectations and cultural norms encourage content moderation. The study concludes that governance reflects interaction between state authority, market forces and society.

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