Governance, Regulation, Legislation, Geostrategies
The 2026 World Development Report focuses on AI capabilities, concentration, and complements
(DigWatch) On 4th August, the World Bank published the 2026 World Development Report: The Promise of Artificial Intelligence. The Report observes that artificial intelligence (AI) is diffusing faster than previous general‑purpose technologies. It states that the steam engine took about 80 years, electricity about 40 years and the internet about 20 years to reach lower‑income countries, whereas middle‑income countries accounted for half of ChatGPT’s global traffic within six months of its launch. The Report describes this rapid diffusion as a ‘historic opportunity for developing countries because AI can greatly expand access to expertise, on a very compressed timetable, to solve complex problems that have defied solutions for decades’, and adds that ‘AI can do in a decade what might otherwise take a century’. – https://dig.watch/updates/the-2026-world-development-report-focuses-on-ai-capabilities-concentration-and-complements
World Bank tells developing countries to adapt AI, not chase frontier models
(DigWatch) Developing countries should adopt existing AI tools and adapt them to local needs rather than investing scarce resources in frontier models, according to the World Bank’s World Development Report 2026. Titled ‘The Promise of Artificial Intelligence’, the report argues that AI could improve public services, expand access to expertise and support economic growth. Lower-cost applications, described as ‘small AI’, can operate through text messages, voice calls and basic mobile phones, making them useful where electricity, connectivity and computing resources remain limited. – https://dig.watch/updates/world-bank-developing-countries-to-embrace-ai
China adds AI to five-year geological hazard prevention plan
(DigWatch) China has launched a five-year action plan to strengthen geological hazard prevention, climate resilience and local emergency-response capabilities between 2026 and 2030. Issued by the Ministry of Natural Resources, the programme includes relocating 50,000 households from high-risk areas and carrying out engineering work at more than 4,200 locations prone to geological disasters. Authorities also plan to address safety risks at over 6,100 additional sites. The plan calls for wider risk surveys, stronger monitoring networks and improved early warning systems. Research will also examine how extreme rainfall, rapid weather changes and sudden transitions between drought and flooding can trigger geological hazards. A new national information platform will connect geological disaster systems operated at national and provincial levels. China plans to use big data, large AI models and other emerging technologies to improve risk analysis, forecasting and emergency response. – https://dig.watch/updates/china-adds-ai-geological-hazard-prevention-plan
Courts and Litigation
Apple renews UK legal fight over encrypted iCloud data
(DigWatch) Apple has launched a new legal challenge against a UK government order reportedly requiring the company to maintain the ability to provide access to encrypted iCloud data belonging to British users. The company filed its complaint with the Investigatory Powers Tribunal in July, challenging the Home Secretary’s powers to issue Technical Capability Notices under the Investigatory Powers Act. The revised notice follows the withdrawal of an earlier order that reportedly extended beyond the UK and included American users. Its precise terms remain secret, but the latest demand is understood to target data protected by Apple’s Advanced Data Protection feature in the UK. – https://dig.watch/updates/apple-challenges-uk-over-encrypted-icloud-data
Security and Surveillance
Exposed SISVISA Database Leaks 102,000 Brazilian Health Surveillance Records
(Pierluigi Paganini – Security Affairs) Researcher Jeremiah Fowler found a publicly accessible database that turned out to belong to SISVISA, Brazil’s Health Surveillance Information System, and alerted ExpressVPN, which later shared the findings with Hackread. The exposed instance held exactly 102,215 files, roughly 79 GB of data, tied to a platform regulators use to track health rules, issue licences and manage inspections for hospitals, restaurants and pharmacies. In other words, it wasn’t some forgotten test box in a corner; it was wired into how the state does its job. “Over 102,000 private records belonging to Brazil’s health surveillance system were left online without passwords or basic encryption. Security researcher Jeremiah Fowler found this publicly accessible database and alerted cybersecurity firm ExpressVPN, which later shared the details with Hackread.com.” reads the report published by Hackread. “According to Fowler, this open database stored exactly 102,215 files (around 79GB). Further probing revealed that these records belong to Brazil’s Health Surveillance Information System (SISVISA). For your information, this is a crucial platform used by Brazilian health authorities to track public health rules, issue business permits, and manage inspections for hospitals, restaurants, and pharmacies.” – https://securityaffairs.com/196766/data-breach/exposed-sisvisa-database-leaks-102000-brazilian-health-surveillance-records.html
Meta AI Model Hacked a Company During Testing, Marking Third AI Lab Incident
(Pierluigi Paganini – Security Affairs) Meta confirmed that one of its AI models breached an unidentified company during cybersecurity testing, after its independent testing partner Irregular gave the model unintended internet access through a misconfiguration. This is the third major AI lab to disclose a testing breach in two weeks: OpenAI’s agent hacked Hugging Face in July, Anthropic disclosed last week that its models compromised three companies, and now Meta. The pattern is no longer a one-off incident. The model “exploited a security vulnerability in a third-party service, in a manner similar to previously reported instances with other companies,” Meta said in a statement, as reported by Reuters. Irregular confirmed the incident was caused by the same evaluation environment misconfiguration previously disclosed by Anthropic, not by a sandbox escape or an advanced cyberattack. – https://securityaffairs.com/196731/security/meta-ai-model-hacked-a-company-during-testing-marking-third-ai-lab-incident.html
Snowflake Hacker Pleads Guilty After Breaching 165 Companies and Stealing Billions of Records
(Pierluigi Paganini – Security Affairs) Connor Riley Moucka, 26, of Kitchener, Ontario, pleaded guilty this week to a computer hacking conspiracy that compromised over 165 organizations, stole billions of customer records, and extorted multiple victims for millions of dollars. “Connor Riley Moucka, 26, of Kitchener, Ontario, pleaded guilty today to a widespread computer hacking conspiracy that resulted in the compromise of over 165 victim organizations, the theft of billions of sensitive customer records and the extortion of numerous victims.” states DoJ. The unnamed “U.S.-based software-as-a-service company” at the center of the scheme is Snowflake, the cloud data platform, though the DOJ press release doesn’t name it directly. Moucka was arrested just six months after the breaches began, which is either impressive law enforcement work or a sign that he wasn’t as careful as he thought. – https://securityaffairs.com/196714/security/snowflake-hacker-pleads-guilty-after-breaching-165-companies-and-stealing-billions-of-records.html
Defence, Warfare
US turns to AI agents for drone production and cyber defence
(DigWatch) The US National Institute of Standards and Technology and Department of Energy have formed a partnership to apply AI to scientific research, manufacturing and critical infrastructure security. A memorandum of understanding allows the agencies to coordinate facilities, technical expertise and research resources under the White House-led Genesis Mission. Areas of cooperation include biotechnology, quantum science and materials discovery. The Genesis Mission aims to double the productivity and impact of US science and engineering within a decade by expanding the use of AI. NIST will focus on technologies intended to move from government research into commercial and industrial applications. – https://dig.watch/updates/nist-doe-genesis-mission-ai-partnership
Frontiers
MIT study warns AI explanations can deepen diagnostic errors
(DigWatch) AI-generated explanations can encourage non-experts to accept incorrect medical recommendations, according to research involving MIT and several partner institutions. Published in Nature Medicine, the study examined how different forms of explainable AI affected 623 members of the public and 153 primary care physicians performing skin-disease diagnosis tasks. Participants received assistance in several formats, including a model prediction with a confidence score, similar medical images, heat maps highlighting relevant areas and explanations generated by a large language model. AI assistance generally improved diagnostic accuracy and reduced performance differences across skin tones. However, the benefits varied significantly according to users’ medical expertise and the accuracy of the model’s recommendation. – https://dig.watch/updates/mit-study-limits-of-explainable-ai-in-healthcare
NVIDIA opens robotaxi AI model for commercial use
(DigWatch) NVIDIA has released Alpamayo 2 Super for commercial use, giving automakers and suppliers access to an openly licensed AI model designed for developing robotaxis and other autonomous vehicles. The 34-billion-parameter model combines perception, reasoning and trajectory planning within a single multimodal system. Built on NVIDIA Cosmos 3 Super Reasoner and post-trained through reinforcement learning, it processes video from cameras positioned around a vehicle to analyse lane changes, merges, unprotected turns and interactions involving multiple road users. – https://dig.watch/updates/nvidia-opens-robotaxi-ai-model-for-commercial-use
NVIDIA joins US initiative to expand regional AI research
(DigWatch) The US National Science Foundation (NSF) has launched a $100 million programme to establish up to 10 state and regional AI infrastructure hubs, with NVIDIA among the technology companies that intend to support participating institutions. The programme will bring together universities, state and local governments, businesses and philanthropic organisations to expand access to computing, data and software for AI-enabled scientific research. NSF expects to make one award per state or multi-state region, with individual grants generally ranging from $4 million to $12 million over five years. – https://dig.watch/updates/nvidia-joins-us-initiative-to-expand-regional-ai-research