A Guardian investigation has found that Google’s AI Overviews have displayed false and misleading health information that could put people at risk of harm. The summaries, which appear at the top of search results, are generated using AI and are presented as reliable snapshots of key information.
The investigation identified multiple cases where Google’s AI summaries provided inaccurate medical advice. Examples included incorrect guidance for pancreatic cancer patients, misleading explanations of liver blood test results, and false information about women’s cancer screening.
Health experts warned that such errors could lead people to dismiss symptoms, delay treatment, or follow harmful advice. Some charities said the summaries lacked essential context and could mislead users during moments of anxiety or crisis.
Concerns were also raised about inconsistencies, with the same health queries producing different AI-generated answers at different times. Experts said this variability undermines trust and increases the risk that misinformation will influence health decisions.
Google said most AI Overviews are accurate and helpful, and that the company continually improves quality, particularly for health-related topics. It said action is taken when summaries misinterpret content or lack appropriate context.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
Face-to-face interviews and oral verification could become a routine part of third-level assessments under new recommendations aimed at addressing the improper use of AI. Institutions are being encouraged to redesign assessment methods to ensure student work is authentic.
The proposals are set out in new guidelines published by the Higher Education Authority (HEA) of Ireland, which regulates universities and other third-level institutions. The report argues that assessment systems must evolve to reflect the growing use of generative AI in education.
While encouraging institutions to embrace AI’s potential, the report stresses the need to ensure students are demonstrating genuine learning. Academics have raised concerns that AI-generated assignments are increasingly difficult to distinguish from original student work.
To address this, the report recommends redesigning assessments to prioritise student authorship and human judgement. Suggested measures include oral verification, process-based learning, and, where appropriate, a renewed reliance on written exams conducted without technology.
The authors also caution against relying on AI detection tools, arguing that integrity processes should be based on dialogue and evidence. They call for clearer policies, staff and student training, and safeguards around data use and equitable access to AI tools.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
India’s government has set out plans to democratise AI infrastructure nationwide. The strategy focuses on expanding access beyond major technology hubs.
Officials aim to increase availability of computing power, datasets and AI models. Startups, researchers and public institutions are key intended beneficiaries.
New initiatives under IndiaAI and national supercomputing programmes will boost domestic capacity. Authorities say local compute access reduces reliance on foreign providers.
Digital public platforms will support data sharing and model development. The approach seeks inclusive innovation across education, healthcare and governance across India.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Growing concern over data privacy and subscription fatigue has led an independent developer to create WitNote, an AI note-taking tool that runs entirely offline.
The software allows users to process notes locally on Windows and macOS rather than relying on cloud-based services where personal information may be exposed.
WitNote supports lightweight language models such as Qwen2.5-0.5B that can run with limited storage requirements. Users may also connect to external models through API keys if preferred.
Core functions include rewriting, summarising and extending content, while a WYSIWYG Markdown editor provides a familiar workflow without network delays, instead of relying on web-based interfaces.
Another key feature is direct integration with Obsidian Markdown files, allowing notes to be imported instantly and managed in one place.
The developer says the project remains a work in progress but commits to ongoing updates and user-driven improvements, even joining Apple’s developer programme personally to support smoother installation.
For users seeking AI assistance while protecting privacy and avoiding monthly fees, WitNote positions itself as an appealing offline alternative that keeps full control of data on the local machine.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Illinois Secretary of State Alexi Giannoulias has warned residents to stay alert for fraudulent text messages claiming unpaid traffic violations or tolls. Officials say the messages are part of a phishing campaign targeting Illinois drivers.
The scam texts typically warn recipients that their vehicle registration or driving privileges are at risk of suspension. The messages urge immediate action via links that steal money or personal information.
The Secretary of State’s office said it sends text messages only to remind customers about scheduled DMV appointments. It does not communicate by text about licence status, vehicle registration issues, or enforcement actions.
Officials advised residents not to click on links or provide personal details in response to such messages. The texts are intended to create fear and pressure victims into acting quickly.
Residents who receive scam messages are encouraged to report them to the Federal Trade Commission through its online fraud reporting system.
Would you like to learn more about AI, tech, and digital diplomacy? If so, ask our Diplo chatbot!
The booming influencer economy of Belgium is colliding with an advertising rulebook that many creators say belongs to another era.
Different obligations across federal, regional and local authorities mean that wording acceptable in one region may trigger a reprimand in another. Some influencers have even faced large fines for administrative breaches such as failing to publish business details on their profiles.
In response, the Influencer Marketing Alliance in Belgium has launched a certification scheme designed to help creators navigate the legal maze instead of risking unintentional violations.
Influencers complete an online course on advertising and consumer law and must pass a final exam before being listed in a public registry monitored by the Jury for Ethical Practices.
Major brands, including L’Oréal and Coca-Cola, already prefer to collaborate with certified creators to ensure compliance and credibility.
Not everyone is convinced.
Some Belgian influencers argue that certification adds more bureaucracy at a time when they already struggle to understand overlapping rules. Others see value as a structured reminder that content creators remain legally responsible for commercial communication shared with followers.
The alliance is also pushing lawmakers to involve influencers more closely when drafting future rules, including taxation and safeguards for child creators.
Consumer groups such as BEUC support clearer definitions and obligations under the forthcoming EU Digital Fairness Act, arguing that influencer advertising should follow the same standards as other media instead of remaining in a grey zone.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
AI dictation reached maturity during the years after many attempts of patchy performance and frustrating inaccuracies.
Advances in speech-to-text engines and large language models now allow modern dictation tools to recognise everyday speech more reliably while keeping enough context to format sentences automatically instead of producing raw transcripts that require heavy editing.
Several leading apps have emerged with different strengths. Wispr Flow focuses on flexibility with style options and custom vocabulary, while Willow blends automation with privacy by storing transcripts locally.
Monologue also prioritises privacy by allowing users to download the model and run transcription entirely on their own machines. Superwhisper caters for power users by supporting multiple downloadable models and transcription from audio or video files.
Other tools take different approaches. VoiceTypr offers an offline-first design with lifetime licensing, Aqua promotes speed and phrase-based shortcuts, Handy provides a simple free open source starting point, and Typeless gives one of the most generous free allowances while promising strong data protection.
Each reflects a wider trend where developers try to balance convenience, privacy, control and affordability.
Users now benefit from cleaner, more natural-sounding transcripts instead of the rigid audio typing tools of previous years. AI dictation has become faster, more accurate and far more usable for everyday note-taking, messaging and work tasks.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
AI tools are increasingly used for simple everyday calculations, yet a new benchmark suggests accuracy remains unreliable.
The ORCA study tested five major chatbots across 500 real-world maths prompts and found that users still face roughly a 40 percent chance of receiving the wrong answer.
Gemini from Google recorded the highest score at 63 percent, with xAI’s Grok almost level at 62.8 percent. DeepSeek followed with 52 percent, while ChatGPT scored 49.4 percent, and Claude placed last at 45.2 percent.
Performance varied sharply across subjects, with maths and conversion tasks producing the best results, but physics questions dragged scores down to an average accuracy below 40 percent.
Researchers identified most errors as sloppy calculations or rounding mistakes, rather than deeper failures to understand the problem. Finance and economics questions highlighted the widest gaps between the models, while DeepSeek struggled most in biology and chemistry, with barely one correct answer in ten.
Users are advised to double-check results whenever accuracy is crucial. A calculator or a verified source is still advised instead of relying entirely on an AI chatbot for numerical certainty.
Would you like to learn more aboutAI, tech and digital diplomacy? If so, ask our Diplo chatbot!
China has proposed stringent new rules for AI aimed at protecting children and preventing chatbots from providing advice that could lead to self-harm, violence, or gambling.
The draft regulations, published by the Cyberspace Administration of China (CAC), require developers to include personalised settings, time limits, and parental consent for services offering emotional companionship.
High-risk chats involving self-harm or suicide must be passed to a human operator, with guardians or emergency contacts alerted. AI providers must not produce content that threatens national security, harms national honour, or undermines national unity.
The rules come as AI usage surges, with platforms such as DeepSeek, Z.ai, and Minimax attracting millions of users in China and abroad. The CAC supports safe AI use, including tools for local culture and elderly companionship.
The move reflects growing global concerns over AI’s impact on human behaviour. Notably, OpenAI has faced legal challenges over alleged chatbot-related harm, prompting the company to create roles focused on tracking AI risks to mental health and cybersecurity.
China’s draft rules signal a firm approach to regulating AI technology as its influence expands rapidly.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
Korean Air has disclosed a data breach affecting about 30,000 employees. Stolen records were taken from systems operated by a former subsidiary.
The breach occurred at catering supplier KC&D, sold off in 2020. Hackers, who had previously attacked the Washington Post accessed employee names and their bank account details, while customer data remained unaffected.
Investigators linked the incident to exploits in Oracle E-Business Suite. Cybercriminals abused zero day flaws during a wider global hacking campaign.
The attack against Korean Air has been claimed by the Cl0p ransomware group. Aviation firms worldwide have reported similar breaches connected to the same campaign.
Would you like to learn more about AI, tech and digital diplomacy? If so, ask our Diplo chatbot!
