Digital identities

AI health tools need clinicians to prevent serious risks, Oxford study warns

The University of Oxford has warned that AI in healthcare, primarily through chatbots, should not operate without human oversight.

Researchers found that relying solely on AI for medical self-assessment could worsen patient outcomes instead of improving access to care. The study highlights how these tools, while fast and data-driven, fall short in delivering the judgement and empathy that only trained professionals can offer.

The findings raise alarm about the growing dependence on AI to fill gaps caused by doctor shortages and rising costs. Chatbots are often seen as scalable solutions, but without rigorous human-in-the-loop validation, they risk providing misleading or inconsistent information, particularly to vulnerable groups.

Rather than helping, they might increase health disparities by delaying diagnosis or giving patients false reassurance.

Experts are calling for safer, hybrid approaches that embed clinicians into the design and ongoing use of AI tools. The Oxford researchers stress that continuous testing, ethical safeguards and clear protocols must be in place.

Instead of replacing clinical judgement, AI should support it. The future of digital healthcare hinges not just on innovation but on responsibility and partnership between technology and human care.

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

Google pushes users to move away from passwords

Google urges users to move beyond passwords, citing widespread reuse and vulnerability to phishing attacks. The company is now promoting alternatives like passkeys and social sign-ins as more secure and user-friendly options.

Data from Google shows that half of users reuse passwords, while the rest either memorise or write them down. Gen Z is leading the shift and is significantly more likely to adopt passkeys and social logins than older generations.

Passkeys, stored on user devices, eliminate traditional password input and reduce phishing risks by relying on biometrics or device PINs for authentication. However, limited app support and difficulty syncing across devices remain barriers to broader adoption.

Google highlights that while social sign-ins offer convenience, they come with privacy trade-offs by giving large companies access to more user activity data. Users still relying on passwords are advised to adopt app-based two-factor authentication over SMS or email, which are far less secure.

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

Fake DeepSeek ads deliver ‘BrowserVenom’ malware to curious AI users

Cybercriminals are exploiting the surge in interest around local AI tools by spreading a new malware strain via Google ads.

According to antivirus firm Kaspersky, attackers use fake ads for DeepSeek’s R1 AI model to deliver ‘BrowserVenom,’ malware designed to intercept and manipulate a user’s internet traffic instead of merely infecting the device.

The attackers purchased ads appearing in Google search results for ‘deep seek r1.’ Users who clicked were redirected to a fake website—deepseek-platform[.]com—which mimicked the official DeepSeek site and offered a file named AI_Launcher_1.21.exe.

Kaspersky’s analysis of the site’s source code uncovered developer notes in Russian, suggesting the campaign is operated by Russian-speaking actors.

Once launched, the fake installer displayed a decoy installation screen for the R1 model, but silently deployed malware that altered browser configurations.

BrowserVenom rerouted web traffic through a proxy server controlled by the hackers, allowing them to decrypt browsing sessions and capture sensitive data, while evading most antivirus tools.

Kaspersky reports confirmed infections across multiple countries, including Brazil, Cuba, India, and South Africa.

The malicious domain has since been taken down. However, the incident highlights the dangers of downloading AI tools from unofficial sources. Open-source models like DeepSeek R1 require technical setup, typically involving multiple configuration steps, instead of a simple Windows installer.

As interest in running local AI grows, users should verify official domains and avoid shortcuts that could lead to malware.

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

Meta sues AI firm over fake nude images created without consent

Meta has filed a lawsuit against Joy Timeline HK Ltd in Hong Kong, accusing the firm of using its platforms to promote a generative AI app called CrushAI.

The app allows users to digitally strip clothes from images of people, often without consent. Meta said the company repeatedly attempted to bypass ad review systems to push harmful content, advertising phrases like ‘see anyone naked’ on Facebook and Instagram.

The lawsuit follows Meta’s broader investigation into ‘nudity’ apps, which are increasingly being used to create sexualised deepfakes. Despite bans on nonconsensual explicit content, the company said such apps evade detection by disguising ads or rotating domain names after bans.

According to research by Cornell Tech, over 8,000 ads linked to CrushAI appeared on Meta platforms in recent months. Meta responded by updating its detection systems with a broader range of flagged terms and emojis.

While many of the manipulated images target celebrities, concerns are growing about the use of such technology to exploit minors. In one case in Florida, two teenagers used similar AI tools to create sexualised images of classmates.

The issue has sparked legal action in the US, where the Take It Down Act, signed into law earlier this year, criminalises the publication of nonconsensual deepfake imagery and simplifies removal processes for victims.

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

AI startup faces lawsuit from Disney and Universal

Two of Hollywood’s most powerful studios, Disney and Universal, have launched a copyright infringement lawsuit against the AI firm Midjourney, accusing it of illegally replicating iconic characters.

The studios claim the San Francisco-based company copied their creative works without permission, describing it as a ‘bottomless pit of plagiarism’.

Characters such as Darth Vader, Elsa, and the Minions were cited in the 143-page complaint, which alleges Midjourney used these images to train its AI system and generate similar content.

Disney and Universal argue that the AI firm failed to invest in the creative process, yet profited heavily from the output — reportedly earning $US300 million in paid subscriptions last year.

Despite early attempts by the studios to raise concerns and propose safeguards already adopted by other AI developers,

Midjourney allegedly ignored them and pressed ahead with further product releases. The company, which calls itself a small, self-funded team of 11, has declined to comment on the lawsuit directly but insists it has a long future ahead.

Disney’s legal chief, Horacio Gutierrez, stressed the importance of protecting creative works that result from decades of investment. While supporting AI as a tool for innovation, he maintained that ‘piracy is piracy’, regardless of whether humans or machines carry it out.

The studios are seeking damages and a court order to stop the AI firm from continuing its alleged copyright violations.

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

Wikipedia halts AI summaries test after backlash

Wikipedia has paused a controversial trial of AI-generated article summaries following intense backlash from its community of volunteer editors.

The Wikimedia Foundation had planned a two-week opt-in test for mobile users using summaries produced by Aya, an open-weight AI model developed by Cohere.

However, the reaction from editors was swift and overwhelmingly negative. The discussion page became flooded with objections, with contributors arguing that such summaries risked undermining the site’s reputation for neutrality and accuracy.

Some expressed concerns that inserting AI content would override Wikipedia’s long-standing collaborative approach by effectively installing a single, unverifiable voice atop articles.

Editors warned that AI-generated summaries lacked proper sourcing and could compromise the site’s credibility. Recent AI blunders by other tech giants, including Google’s glue-on-pizza mishap and Apple’s false death alert, were cited as cautionary examples of reputational risk.

For many, the possibility of similar errors appearing on Wikipedia was unacceptable.

Marshall Miller of the Wikimedia Foundation acknowledged the misstep in communication and confirmed the project’s suspension.

While the Foundation remains interested in exploring AI to improve accessibility, it has committed to ensuring any future implementation involves direct participation from the Wikipedia community.

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

INTERPOL cracks down on global cybercrime networks

Over 20,000 malicious IP addresses and domains linked to data-stealing malware have been taken down during Operation Secure, a coordinated cybercrime crackdown led by INTERPOL between January and April 2025.

Law enforcement agencies from 26 countries worked together to locate rogue servers and dismantle criminal networks instead of tackling threats in isolation.

The operation, supported by cybersecurity firms including Group-IB, Kaspersky and Trend Micro, led to the removal of nearly 80 per cent of the identified malicious infrastructure. Authorities seized 41 servers, confiscated over 100GB of stolen data and arrested 32 suspects.

More than 216,000 individuals and organisations were alerted, helping them act quickly by changing passwords, freezing accounts or blocking unauthorised access.

Vietnamese police arrested 18 people, including a group leader found with cash, SIM cards and business records linked to fraudulent schemes. Sri Lankan and Nauruan authorities carried out home raids, arresting 14 suspects and identifying 40 victims.

In Hong Kong, police traced 117 command-and-control servers across 89 internet providers. INTERPOL hailed the effort as proof of the impact of cross-border cooperation in dismantling cybercriminal infrastructure instead of allowing it to flourish undisturbed.

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

Massive leak exposes data of millions in China

Cybersecurity researchers have uncovered a brief but significant leak of over 600 gigabytes of data, exposing information on millions of Chinese citizens.

The haul, containing WeChat, Alipay, banking, and residential records, is part of a centralised system, possibly aimed at large-scale surveillance instead of a random data breach.

According to research from Cybernews and cybersecurity consultant Bob Diachenko, the data was likely used to build individuals’ detailed behavioural, social and economic profiles.

They warned the information could be exploited for phishing, fraud, blackmail or even disinformation campaigns instead of remaining dormant. Although only 16 datasets were reviewed before the database vanished, they indicated a highly organised and purposeful collection effort.

The source of the leak remains unknown, but the scale and nature of the data suggest it may involve government-linked or state-backed entities rather than lone hackers.

The exposed information could allow malicious actors to track residence locations, financial activity and personal identifiers, placing millions at risk instead of keeping their lives private and secure.

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

Digital Social Security cards coming this summer

The US Social Security Administration is launching digital access to Social Security numbers in the summer of 2025 through its ‘My Social Security’ portal. The initiative aims to improve convenience, reduce physical card replacement delays, and protect against identity theft.

The digital rollout responds to the challenges of outdated paper cards, rising fraud risks, and growing demand for remote access to US government services. Cybersecurity experts also recommend using VPNs, antivirus software, and identity monitoring services to guard against phishing scams and data breaches.

While it promises faster and more secure access, experts urge users to bolster account protection through strong passwords, two-factor authentication, and avoidance of public Wi-Fi when accessing sensitive data.

Users should regularly check their credit reports and SSA records and consider requesting an IRS PIN to prevent tax-related fraud. The SSA says this move will make Social Security more efficient without compromising safety.

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

Reddit targets AI firm over scraped sports posts

Reddit has taken legal action against AI company Anthropic, accusing it of scraping content from the platform’s sports-focused communities.

The lawsuit claims Anthropic violated Reddit’s user agreement by collecting posts without permission, particularly from fan-driven discussions that are central to how sports content is shared online.

Reddit argues the scraping undermines its obligations to over 100 million daily users, especially around privacy and user control. According to the filing, Anthropic’s actions override assurances that users can manage or delete their content as they see fit.

The platform emphasises that users gain no benefit from technology built using their contributions.

These online sports communities are rich sources of original fan commentary and analysis. On a large scale, such content could enable AI models to imitate sports fan behaviour with impressive accuracy.

While teams or platforms might use such models to enhance engagement or communication, Reddit warns that unauthorised use brings serious ethical and legal risks.

The case could influence how AI companies handle user-generated content across the internet, not just in sports. As web scraping grows more common, the outcome of the dispute may shape future standards for AI training practices and online content rights.

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