Interdisciplinary approaches

Millions of customer records stolen in Kering luxury brand data breach

Kering has confirmed a data breach affecting several of its luxury brands, including Gucci, Balenciaga, Brioni, and Alexander McQueen, after unauthorised access to its Salesforce systems compromised millions of customer records.

Hacking group ShinyHunters has claimed responsibility, alleging it exfiltrated 43.5 million records from Gucci and nearly 13 million from the other brands. The stolen data includes names, email addresses, dates of birth, sales histories, and home addresses.

Kering stated that the incident occurred in June 2025 and did not compromise bank or credit card details or national identifiers. The company has reported the breach to the relevant regulators and is notifying the affected customers.

Evidence shared by ShinyHunters suggests Balenciaga made an initial ransom payment of €500,000 before negotiations broke down. The group released sample data and chat logs to support its claims.

ShinyHunters has exploited Salesforce weaknesses in previous attacks targeting luxury, travel, and financial firms. Questions remain about the total number of affected customers and the potential exposure of other Kering brands.

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

Codex gets smarter with GPT-5, targets GitHub Copilot and rivals

OpenAI has optimised its new GPT-5 model for Codex, its agentic software development tool, boosting performance on both quick coding sessions and long, complex projects. CEO Sam Altman said Codex already accounts for 40% of platform traffic.

GPT-5 Codex can now build full projects, add features, run tests, refactor large codebases, and conduct detailed code reviews. It dynamically adjusts the time spent ‘thinking’ based on task complexity, allowing both interactive pair programming and extended autonomous work.

OpenAI stated that the model can run independently for over seven hours, completing refactorings, fixing test failures, and delivering finished code. Early tests indicate that it catches critical bugs more reliably, allowing developers to focus on the most important issues.

The upgraded Codex is available via terminal, IDE integrations, the web, and GitHub, and comes bundled with ChatGPT Plus, Pro, Business, Edu, and Enterprise subscriptions. OpenAI launched Codex CLI in April and a research preview in May.

With GPT-5 Codex, OpenAI aims to capture market share from GitHub Copilot, Google’s Gemini, Anthropic’s Claude, and startups such as Anysphere and Windsurf. The company claims the new version delivers faster, higher-quality results for developers at every stage of the software lifecycle.

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

PDGrapher AI tool aims to speed up precision medicine development

Harvard Medical School researchers have developed an AI tool that could transform drug discovery by identifying multiple drivers of disease and suggesting treatments to restore cells to a healthy state.

The model, called PDGrapher, utilises graph neural networks to map the relationships between genes, proteins, and cellular pathways, thereby predicting the most effective targets for reversing disease. Unlike traditional approaches that focus on a single protein, it considers multiple factors at once.

Trained on datasets of diseased cells before and after treatment, PDGrapher correctly predicted known drug targets and identified new candidates supported by emerging research. The model ranked potential targets up to 35% higher and worked 25 times faster than comparable tools.

Researchers are now applying PDGrapher to complex diseases such as Parkinson’s, Alzheimer’s, and various cancers, where single-target therapies often fail. By identifying combinations of targets, the tool can help overcome drug resistance and expedite treatment design.

Senior author Marinka Zitnik said the ultimate goal is to create a cellular ‘roadmap’ to guide therapy development and enable personalised treatments for patients. After further validation, PDGrapher could become a cornerstone in precision medicine.

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

Privacy-preserving AI gets a boost with Google’s VaultGemma model

Google has unveiled VaultGemma, a new large language model built to offer cutting-edge privacy through differential privacy. The 1-billion-parameter model is based on Google’s Gemma architecture and is described as the most powerful differentially private LLM to date.

Differential privacy adds mathematical noise to data, preventing the identification of individuals while still producing accurate overall results. The method has long been used in regulated industries, but has been challenging to apply to large language models without compromising performance.

VaultGemma is designed to eliminate that trade-off. Google states that the model can be trained and deployed with differential privacy enabled, while maintaining comparable stability and efficiency to non-private LLMs.

This breakthrough could have significant implications for developers building privacy-sensitive AI systems, ranging from healthcare and finance to government services. It demonstrates that sensitive data can be protected without sacrificing speed or accuracy.

Google’s research teams say the model will be released with open-source tools to help others adopt privacy-preserving techniques. The move comes amid rising regulatory and public scrutiny over how AI systems handle personal data.

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

EdChat AI app set for South Australian schools amid calls for careful use

South Australian public schools will soon gain access to EdChat, a ChatGPT-style app developed by Microsoft in partnership with the state government. Education Minister Blair Boyer said the tool will roll out next term across public high schools following a successful trial.

Safeguards have been built into EdChat to protect student data and alert moderators if students type concerning prompts, such as those related to self-harm or other sensitive topics. Boyer said student mental health was a priority during the design phase.

Teachers report that students use EdChat to clarify instructions, get maths solutions explained, and quiz themselves on exam topics. Adelaide Botanic High School principal Sarah Chambers described it as an ‘education equaliser’ that provides students with access to support throughout the day.

While many educators in Australia welcome the rollout, experts warn against overreliance on AI tools. Toby Walsh of UNSW said students must still learn how to write essays and think critically, while others noted that AI could actually encourage deeper questioning and analysis.

RMIT computing expert Michael Cowling said generative AI can strengthen critical thinking when used for brainstorming and refining ideas. He emphasised that students must learn to critically evaluate AI output and utilise the technology as a tool, rather than a substitute for learning.

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

Rising data centre demand pushes utilities to invest

US electricity prices are rising as the energy demands of data centres surge, driven by the rapid growth of AI technologies. The average retail price per kilowatt-hour increased by 6.5% between May 2024 and May 2025, with some states experiencing significantly sharper increases.

Maine saw the sharpest rise in electricity prices at 36.3%, with Connecticut and Utah following closely behind. Utilities are passing on infrastructure costs, including new transmission lines, to consumers. In Northern Virginia, residents could face monthly bill increases of up to $37 by 2040.

Analysts warn that the shift to generative AI will lead to a 160% surge in energy use at data centres by 2030. Water use is also rising sharply, as Google reported its facilities consumed around 6 billion gallons in 2024 alone, amid intensifying global AI competition.

Tech giants are turning to alternative energy to keep pace. Google has announced plans to power data centres with small nuclear reactors through a partnership with Kairos Power, while Microsoft and Amazon are ramping up nuclear investments to secure long-term supply.

President Donald Trump has pledged more than $92 billion in AI and energy infrastructure investments, underlining Washington’s push to ensure the US remains competitive in the AI race despite mounting strain on the grid and water resources.

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

Ukraine urges ethical use of AI in education

AI can help build individual learning paths for Ukraine’s 3.5 million students, but its use must remain ethical, First Deputy Minister of Education and Science Yevhen Kudriavets has said.

Speaking to UNN, Kudriavets stressed that AI can analyse large volumes of information and help students acquire the knowledge they need more efficiently. He said AI could construct individual learning trajectories faster than teachers working manually.

He warned, however, that AI should not replace the educational process and that safeguards must be found to prevent misuse.

Kudriavets also said students in Ukraine should understand the reasons behind using AI, adding that it should be used to achieve knowledge rather than to obtain grades.

The deputy minister emphasised that technology itself is neutral, and how people choose to apply it determines whether it benefits education.

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

YouTube expands AI dubbing to millions of creators

Real-time translation is becoming a standard feature across consumer tech, with Samsung, Google, and Apple all introducing new tools. Apple’s recently announced Live Translation on AirPods demonstrates the utility of such features, particularly for travellers.

YouTube has joined the trend, expanding its multi-language audio feature to millions of creators worldwide. The tool enables creators to add dubbed audio tracks in multiple languages, powered by Google’s Gemini AI, replicating tone and emotion.

The feature was first tested with creators like MrBeast, Mark Rober, and Jamie Oliver. YouTube reports that Jamie Oliver’s channel saw its views triple, while over 25% of the watch time came from non-primary languages.

Mark Rober’s channel now supports more than 30 languages per video, helping creators reach audiences far beyond their native markets. YouTube states that this expansion should make content more accessible to global viewers and increase overall engagement.

Subtitles will still be vital for people with hearing difficulties, but AI-powered dubbing could reduce reliance on them for language translation. For creators, it marks a significant step towards making content truly global.

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

NATO and Seoul expand cybersecurity dialogue and defence ties

South Korea and NATO have pledged closer cooperation on cybersecurity following high-level talks in Seoul this week, according to Yonhap News Agency.

The discussions, led by Ambassador for International Cyber Affairs Lee Tae Woo and NATO Assistant Secretary General Jean-Charles Ellermann-Kingombe, focused on countering cyber threats and assessing risks in the Indo-Pacific and Euro-Atlantic regions.

Launched in 2023, the high-level cyber dialogue aims to deepen collaboration between South Korea and NATO in the cybersecurity domain.

The meeting followed talks between Defence Minister Ahn Gyu-back and NATO Military Committee chair Giuseppe Cavo Dragone during the Seoul Defence Dialogue earlier this week.

Dragone said cooperation would expand across defence exchanges, information sharing, cyberspace, space, and AI as ties between Seoul and NATO strengthen.

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

Qwen3-Next strengthens Alibaba’s position in global AI race

Alibaba has open-sourced its latest AI model, Qwen3-Next, claiming it is ten times more powerful and cheaper to train than its predecessor.

Developed by Alibaba Cloud, the 80-billion-parameter model reportedly performs on par with the company’s flagship Qwen3-235B-A22B while remaining optimised for deployment on consumer-grade hardware.

Qwen3-Next introduces innovations such as hybrid attention for long text processing, high-sparsity mixture-of-experts architecture, and multi-token prediction strategies. These upgrades boost both efficiency and model stability during training.

Alibaba also released Qwen3-Next-80B-A3B-Thinking, a reasoning-focused model that outperformed its own Qwen3-32B-Thinking and Google’s Gemini-2.5-Flash-Thinking in benchmark tests.

The release strengthens Alibaba’s position as a major player in open-source AI, following last week’s preview of its 1-trillion-parameter Qwen-3-Max model, which ranked sixth on UC Berkeley’s LMArena leaderboard.

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