Governance, Regulation, Legislation, Geostrategies

IEA outlines policies to manage AI data centre energy demand – The International Energy Agency (IEA) has published a report examining how governments and regulators can integrate rapidly expanding data centres into electricity systems while addressing energy affordability, sustainability and security. Prepared at the request of France’s G7 Presidency, Powering Artificial Intelligence reviews policies and regulatory approaches across 34 jurisdictions representing approximately 90% of global data centre capacity. The agency identified 134 policies and regulations governing data centre integration, nearly 40% of which were announced or implemented over the preceding 12 months. The report highlights growing pressure on electricity networks as demand for AI computing infrastructure increases. According to the IEA, global data centre electricity consumption rose by 17% in 2025 and could double over the following five years. Governments are responding through measures to improve demand forecasting, reform grid connection queues and prioritise viable projects. Some jurisdictions are also introducing flexible connection arrangements or requiring data centres to contribute to electricity supply and grid stability. The IEA estimates that non-firm connections, which allow electricity demand to be restricted during periods of network stress, could unlock an additional 750–900 gigawatts of grid hosting capacity worldwide. – https://dig.watch/updates/iea-policies-ai-data-centre-energy-demand

EDPB adopts smart glasses data protection statement – The European Data Protection Board (EDPB) has adopted Statement 1/2026 on smart glasses, setting out preliminary guidance on how EU data protection law applies to increasingly common camera- and microphone-equipped eyewear. The Board says the technology creates significant privacy and data protection risks because personal data can be processed in ways users and bystanders find hard to perceive or control. The EDPB defines smart glasses broadly as wearable devices that may incorporate cameras, microphones, speakers, applications and AI systems, including tools for translation or object identification. It warns that their discreet appearance can make recording hard to detect and may facilitate unauthorized collection of images, sound, or conversations. Those concerns echo earlier privacy concerns about smart glasses and wearable devices that can capture people without their knowledge. – https://dig.watch/updates/edpb-smart-glasses-data-protection-statement

 

ITU published report on education for tomorrow – As a part of foresight brief series, the Telecommunication Development Bureau under the International Telecommunication Union (ITU) has published a report titled ‘Shaping education for tomorrow’. It outlines key drivers of change in the global education landscape such as government policies, economy and funding, legal frameworks, social‑cultural‑ethical values, population demographics, impact of the economy on the population, adoption of emerging technologies and technological infrastructure – each assessed for impact and uncertainty. – https://dig.watch/updates/itu-releases-brief-shaping-education-for-tomorrow

Australia introduces Online Safety Bill with user empowerment tools requirement – Australia’s Online Safety Amendment (Digital Duty of Care) Bill 2026 was introduced to the House of Representatives on 15 October 2026. The Bill proposes that a person responsible for an online service has a digital duty of care to ensure a safe online environment in which persons in Australia are protected from seriously harmful material and conduct, and children are protected from material and conduct that is harmful to children. In particular, the Bill places an obligation on online service providers to manage the design features of their service effectively and to provide user‑empowerment tools as required by the Minister through a legislative instrument. A user‑empowerment tool is defined as a tool that allows a user of an online service to manage the way design features of the service operate for the user, including, for example, by providing control over the kind of content that is recommended for the user. – https://dig.watch/updates/australia-introduce-law-for-user-empowerment-tool

Singapore plans guidance for autonomous robotic systems – The Cyber Security Agency of Singapore (CSA) will develop new guidance to help organisations secure autonomous robotic systems throughout their lifecycle as AI increasingly moves from software into machines operating in the physical world. Senior Minister of State Tan Kiat How announced the planned guidance for autonomous robotic systems during the Singapore International Cyber Week Securing Smart Cities event on 13 October. He said the framework would cover systems from design and procurement through testing, deployment, operations, monitoring and updates. – https://dig.watch/updates/autonomous-robotic-systems-singapore-guidance

 

India’s Niti Aayog publishes data centre evaluation framework – The Frontier Tech Hub of the Government of India’s apex policy think tank, NITI Aayog, published a Data Centre Evaluation Framework on 14 October 2026 to help Indian states assess proposed investments. It provides common assessment principles without creating additional regulation or prescribing a uniform approach. The framework offers a common set of parameters for evaluating data centre proposals in India and is organised around five pillars – power and energy; water and cooling; siting and environmental sustainability; economic and strategic value; and policy, governance and implementation. For each pillar it outlines detailed criteria, recommended minimum approval documents and the respective roles of the central government, which is to provide national standards, model guidance and templates, and the state governments, which are to coordinate infrastructure, set conditions and monitor delivery. The guidance is presented as a decision support tool rather than a prescriptive regulation and includes proposals for a digital single window mechanism and milestone‑based monitoring. – https://dig.watch/updates/niti-aayog-data-centre-evaluation-framework

Greece prepares AI safeguards for platform workers – Greece’s Ministry of Labour and Social Security is continuing consultations on draft legislation intended to regulate the use of artificial intelligence (AI) and automated systems in the management of digital platform workers. Labour Minister Niki Kerameus met representatives of national social partners on 13 October to discuss proposals prepared by a legislative committee in connection with the EU’s Platform Work Directive (EU) 2024/2831. According to the ministry, several comments from social partners have already been accepted for incorporation into the draft. – https://dig.watch/updates/greece-social-partners-ai-rules-platform-work

Anthropic is letting METR researchers look under the hood with employee-level access – Anthropic is giving employee-level access to six researchers from METR, a nonprofit that tests frontier AI models, to independently assess Anthropic’s models, safety practices, and risk management, the company said in a blog post on Thursday. The arrangement comes at a time where incidents of AI agents acting outside their instructions have been disclosed every week or two since late August. The incidents have sharpened calls for independent checks on AI companies. “We want to be publicly accountable for the decisions we’re making,” Anthropic said in the blog post. “We aim to close the gap between what evaluators understand about a frontier AI company’s systems and what the company itself understands.” – https://techcrunch.com/2026/10/15/anthropic-is-letting-metr-researchers-look-under-the-hood-with-employee-level-access/

Courts and Litigation

Eight US states file lawsuits alleging OpenAI misled consumers over ChatGPT safety – Eight US states – Iowa, Kansas, Nebraska, West Virginia, Arkansas, Arizona, Idaho, and Montana – each filed a separate lawsuit in their state courts on Wednesday, alleging that OpenAI misled consumers over ChatGPT safety and that the chatbot is an unsafe product that poses risks to children and vulnerable users. The Arkansas lawsuit for instance, accuses the company of violating Arkansas’ Deceptive Trade Practice Act and the state’s Children and Teens’ Online Privacy Protection Act, a 2025 law that established privacy protections for teens online. In particular, the filing alleges that the defendants engage in false or deceptive acts/practices including designing and offering ChatGPT with inadequate safeguards. – https://dig.watch/updates/us-states-file-suits-alleging-openai-misled-users

Security and Surveillance

 

IMF and World Bank Group launched Bangkok Blueprint – The International Monetary Fund (IMF) and the World Bank Group launched the Bangkok Blueprint, a high‑level framework developed in coordination with the Bank of Thailand, during the 2026 IMF–WBG Annual Meetings, with the objective to help countries prevent, detect, and address the growing threat of cyber-enabled fraud in digital financial services. The blueprint recognise fraud schemes often involve social media, digital platforms, e-commerce, and telecommunications networks, sets out twelve key considerations arranged under five pillars: establishing strategic and legal foundations, strengthening the fraud‑mitigation life cycle, leveraging existing risk‑mitigation and enforcement frameworks, expanding domestic and cross‑border cooperation and promoting consumer protection and resilience. It further emphasise clear accountability and timely cooperation across the public and private sectors, both domestically and across borders, while encouraging countries to tailor these considerations to their legal, institutional, and market conditions. – https://dig.watch/updates/img-world-bank-bangkok-blueprint-cyber-fraud

France: Internet-Exposed Danfoss Controllers Drop 71% After Cyber-Attack – Internet-exposed refrigeration controllers in France are disappearing from public view at a dramatic rate following a cyber incident targeting supermarket food supplies. Censys researchers observed a 71% drop in internet-exposed Danfoss StoreView controllers in just three days after the October 11 cyber-attack targeting hundreds of supermarkets in France, suggesting that operators are moving quickly to limit their systems’ online exposure. Danfoss StoreView was among the equipment targeted by this cyber-attack. StoreView instances are typically used as control panels for Danfoss refrigeration controllers that are installed in many French supermarket fridges and freezers. Danfoss, a Danish temperature regulation systems provider, has confirmed its equipment was targeted by the cyber-attack, and said most affected controllers were either older models that no longer received security updates (AK-SM800 and AK-SC355) or newer models running outdated firmware (AK-SM800A). Danfoss patched three vulnerabilities in its AK-SM 800A controllers in March 2025 (CVE-2025-41450) and August 2025 (CVE-2025-41451 and CVE-2025-41452). A year later, a researcher at industrial security provider Claroty demonstrated how CVE-2025-41450 and CVE-2025-41451 could be chained together to achieve remote code execution and thus disable cooling while falsifying temperature readings, potentially concealing failures in the cold chain. While Danfoss has not confirmed the exact initial access vector, a plausible explanation is that attackers exploited unpatched vulnerabilities in internet-exposed Danfoss controllers to gain remote access to supermarket refrigeration systems. – https://www.infosecurity-magazine.com/news/france-expose-danfoss-controllers/

Suspected Kazakhstan-linked hackers target governments across former Soviet states– A cyber-espionage group with suspected ties to Kazakhstan has targeted government agencies across Central Asia and other former Soviet states, seeking information on diplomatic affairs, economic policy and strategic resources, according to new research. The hackers, tracked as UAT-6275 by Cisco Talos and TAG-193 by Recorded Future’s Insikt Group, have conducted at least eight campaigns since January 2025, the two teams said in jointly issued reports. The Record is an editorially independent unit of Recorded Future. The campaign targeted or showed interest in government institutions across nine countries, including Russia, Belarus, Armenia, Uzbekistan and Tajikistan, the researchers said. Researchers said the hackers appeared particularly interested in government agencies responsible for foreign affairs, economic development, agriculture and environmental management, as well as information about regional energy supplies and water resources. “These organizations hold information that can inform negotiations, economic planning, and regional security decisions,” the Insikt Group said. “Across Central Asia, shared river systems and interconnected energy and transport infrastructure give such information particular strategic value.” – https://therecord.media/suspected-kazakhstan-phishing-central-asia-espionage

Microsoft Disrupts Global Yam Store Network Behind One Billion Phishing Messages – Microsoft has teamed up with industry and law enforcement partners to disrupt a major phishing infrastructure network which has targeted tens of thousands of organizations worldwide since January. In a report published on October 14, Microsoft detailed how its work with Japanese police soon unravelled a network of Chinese-language phishing-as-a-service (PhaaS) kit operators, developers, and a marketplace selling tools and infrastructure. “Threat actors purchased and used these tools and infrastructure, including phishing kits and access to compromised or fraudulent Microsoft or Google cloud environments and related credentials, to target organizations across the healthcare, financial services, wholesale distribution, and IT sectors in Japan, the US, Australia, and India,” it explained. The phishing kits alone have been responsible for over a billion phishing emails and texts sent to over 75,000 organizations since January 2026, it added. – https://www.infosecurity-magazine.com/news/microsoft-yam-store-one-billion/

Warning Over High-Risk Vulnerabilities in Widely Used LLM Gateway Tool – High-risk cybersecurity vulnerabilities have been uncovered in a widely deployed large language model (LLM) gateway tool used by organizations to connect their AI applications to AI software providers including OpenAI, Google and Anthropic. With almost 100,000 downloads a month, LiteLLM is an open-source tool widely used across enterprises to aid with the building of AI applications and services. The two vulnerabilities, tracked as CVE-2026-35029 and CVE-2026-59823, could allow any user to gain administrative access to the LLM gateway. According to cybersecurity researchers at Check Point, this could potentially allow an attacker to perform a full system takeover, providing them with file access, the ability to steal login credentials and even the powers to remotely execute code. The findings were detailed in a blog post published on October 13, and the vulnerabilities affect versions of LiteLLM that are older than version 1.83.9. LLM gateways are systems that sit between an application and one or more LLM providers. Instead of calling model APIs directly, applications route requests through a gateway, which then forwards those requests to the appropriate model backend. – https://www.infosecurity-magazine.com/news/high-risk-vulnerabilities-llm-tool/

Hirudo Android Trojan Replaces Banking Apps to Steal Credentials – ThreatFabric researchers detailed a new Android banking trojan called Hirudo that takes a genuinely different approach to defeating banking security. Instead of throwing a fake login screen over the real app, it uninstalls the real app and puts a modified copy in its place. “What sets it apart is the lengths its operators go to once a bank is in their sights: rather than overlaying a fake screen on top of the genuine banking app, Hirudo attempts to replace the banking app itself with a modified copy that quietly disables the bank’s own security checks from the inside.” reads the analysis published by ThreatFabric. Eight Indonesian banking apps from seven banks already have specially designed replacements ready. The malware’s credential-stealing code also targets users in Indonesia, China, India, Mexico, Morocco, and Thailand. Hirudo is delivered as a second app hidden inside a convincing dropper. The dropper looks like a legitimate shopping app, short drama platform, or adult video service. It loads real content through a WebView to make the app look genuine. If it detects a banking app on the device and finds that Hirudo is not yet installed, it displays a fake in-app update prompt. – https://securityaffairs.com/200970/malware/hirudo-android-trojan-replaces-banking-apps-to-steal-credentials.html

New Aoki Infostealer Targets AI Coding Assistant Credentials – New versions of the Aoki infostealer are targeting credential and session files used by popular AI coding assistants, a new report by Symantec has warned. This represents a significant development from to earlier versions of Aoki, which focused on browser, cryptocurrency wallet and password manager data, and did not target AI-related files. Now, the stealer is also able to access API keys and tokens from a number of AI coding assistants, including Anthropic’s Claude, OpenAI’s Codex, SpaceX’s Cursor and Google’s Gemini. The Symantec researchers said that this information can allow a threat actor to carry out malicious activity under a legitimate developer’s identity and API quota, bypassing safeguards that assume requests originate from a trusted user. In addition, the tokens can expose connected repositories, cloud services and continuous integration and continuous delivery (CI/CD) pipelines linked to that account. – https://www.infosecurity-magazine.com/news/aoki-infostealer-ai-coding/

ICO Reprimands Met Over Third-Party Data Leak – The UK’s data protection regulator has issued a reprimand and enforcement notice to the capital’s police force after a string of governance failings related to its use of a third-party platform. The investigation by the Information Commissioner’s Office (ICO) was focused on Metropolitan Police Service (MPS) use of the Auror retail crime intelligence-sharing platform, which it used in a crackdown on shoplifting. The watchdog found that, between March 2023 and September 2024, MPS officers and staff used the platform despite it not being formally approved. Some 500 MPS accounts were created, of which 375 were active, it said in its writeup of the case. – https://www.infosecurity-magazine.com/news/ico-reprimands-met-over-thirdparty/

Defence, Military

 

Directed-energy weapon that can neutralize threats in seconds integrated on unmanned ground vehicle – A San Francisco-based company has showcased that laser-based counter-drone weapon can easily offer aerial protection for autonomous military logistics vehicles. Aurelius Systems recently mounted its Archimedes autonomous laser weapon on an American Rheinmetall unmanned ground vehicle at AUSA 2026. Aurelius Systems revealed that by combining high-power directed energy lasers, advanced sensors, and AI autonomy, Archimedes tracks, targets, and neutralizes aerial threats in seconds. When deployed in numbers, Archimedes units form a resilient defense grid, creating persistent protection for forces, facilities, and critical infrastructure. – https://interestingengineering.com/military/directed-energy-weapon-can-neutralize-aerial-threats

 

US Military drones tap 150,000 radios in push to beat battlefield jamming threats – L3Harris Technologies and Skydio are integrating military drones with tactical radio technology to help U.S. forces maintain communications under electronic warfare threats. The partnership combines Skydio’s X10D drone with L3Harris radios and specialized waveforms designed to protect command links in contested environments. The integration could help American troops operate uncrewed aircraft systems (UAS) when adversaries attempt to disrupt battlefield communications. It also connects drone operations with tactical radio networks already deployed across U.S. and allied forces. The X10D is a compact drone designed for military reconnaissance, giving troops an aerial view of potential threats without exposing personnel to the same risks. – https://interestingengineering.com/military/military-drones-anti-jamming-radio-tech

 

European officials weigh mix of high-tech drones and heavy armor – European governments are encouraging their defense industries to jointly develop and produce unmanned systems with Ukrainian companies, while cautioning to simultaneously new keep tanks and armored vehicles on hand for defending NATO’s eastern flank, officials said at the Warsaw Security Forum. Dilan Yeşilgöz-Zegerius, the deputy prime minister and defense minister of the Netherlands, highlighted the Dutch defense sector’s ties with Ukraine in the field of drones. “Our armed forces are really training, as we speak, every day with the knowledge the Ukrainians have, because … they have the most innovative Army in Europe,” Yeşilgöz-Zegerius said on Oct. 12. “And what we are doing at the same time … is investing in our own defense industry. Because it was also, of course, due to a lack of spending, not that strong anymore. And in Europe we have that issue”. Meanwhile, while investing in drone and counter-drone systems, NATO allies bordering Russia are also procuring heavy gear for their land forces and seeking to secure a significant share of manufacturing work for their domestic defense industries. Lithuanian Defense Minister Robertas Kaunas told Defense News at the forum that the country’s planned acquisition of six-wheel-drive armored personnel carriers from Finland’s Patria is an example of how allies can work together on major purchases for their land forces. The vehicles are to support the Baltic state’s National Division, its main combat unit. – https://www.defensenews.com/global/europe/2026/10/14/european-officials-weigh-mix-of-high-tech-drones-and-heavy-armor/

 

US Army taps 9 companies to build next-gen command and control software – The U.S. Army announced that it is awarding initial contracts to several defense technology companies to build applications for its new battlefield software. The service will pay a one-year total of $93.6 million, spread out across nine companies, to create applications for its Next Generation Command and Control, or NGC2, software. “While the Army is providing capability we can deploy today, we’re also providing a blueprint and reference point for how industry can integrate in the future with interoperability and speed,” said Brig. Gen. Shane Taylor, portfolio management executive for the Command and Control Information Network. The NGC2 will provide a shared network for technologies, weapons, sensors, devices and applications to operate within that will increase decision-making speed on the battlefield. The chosen companies include General Dynamics Mission Systems, Air Space Intelligence Federal, Immersive Wisdom, LMI, Mente Systems, Stilman Advanced Strategies, Onebrief, Rune Technologies and AIR. – https://www.defensenews.com/industry/techwatch/2026/10/14/us-army-taps-9-companies-to-build-next-gen-command-and-control-software/

Frontiers

World’s first fast-spectrum molten chloride reactor experiment nears critical testing – Chloride Reactor Experiment (MCRE) aims to test a fast-spectrum molten chloride salt reactor under sustained fission conditions. Its results could help engineers move the technology closer to commercial use. The DOE has approved additional cost-shared funding to accelerate the experiment’s final development stage. Southern Company, Idaho National Laboratory and CORE POWER are collaborating on the project through the department’s Advanced Reactor Demonstration Program. – https://interestingengineering.com/energy/fast-spectrum-molten-chloride-reactor-test

 

New nuclear waste sensor detects 100 kHz vibrations, could help spot canister cracks – Researchers at the University of Pittsburgh have demonstrated a fiber-optic sensing system capable of detecting acoustic signals above 100 kilohertz (kHz), roughly five times the frequency cited for conventional systems, in a step toward monitoring cracks and corrosion in nuclear fuel storage canisters. Developed with Pittsburgh-based startup Sensible Photonics, the system underwent field testing at Idaho National Laboratory (INL) in September. The researchers used a mock nuclear fuel canister to test whether the sensors could detect vibrations and locate their source. The technology is designed to monitor steel canisters that hold spent nuclear fuel. It could eventually be retrofitted onto existing storage casks, allowing operators to detect acoustic signals associated with developing cracks before they compromise the containers. The researchers conducted the tests from September 8 to 11, using different sensor configurations to assess the system’s sensitivity. The demonstration marked the completion of the project’s first phase, supported by funding from the US Department of Energy. – https://interestingengineering.com/energy/fiber-optic-sensor-nuclear-waste-casks-test

 

Meta launches an open-weight AI model that runs on Macs – Meta has launched what it calls the “most capable” open-weight model in the United States. Dubbed Muse Spark Open, the model also has the distinction of having been designed to work with Macs. The new model, which leverages 816 billion parameters and a 1 million token context window, claims to offer superior performance to other popular open-weight models, according to evaluation results released by Meta on Thursday. Muse Spark Open is natively multimodal and built for complex coding tasks and agentic labor. It has also been designed to run locally on Apple’s computers. However, you will still need a lot of power to run the model locally, with a requirement of at least 1 terabyte of RAM, per Engadget. The model will run best via the Mac Studio that comes equipped with the M5 Ultra chip. – https://techcrunch.com/2026/10/15/meta-launches-an-open-weight-model-that-runs-on-macs/

 

Amazon devices chief Panos Panay sees a future where AI devices replace the need to use apps – Amazon is placing its bets on ambient computing, not apps, as the future of personal AI. Alongside the launch of the company’s new AI wearables from Bee, Panos Panay, who serves as Amazon’s SVP of Devices, Alexa & Leo, took the stage at TechCrunch Disrupt on Thursday to talk more in depth about where he sees personal computing headed. “A lot of times, what you’re doing right now — what we all do right now — is we have to remember what we’re asking our tool to do for us,” Panay began, explaining that this is how we engage with our modern smartphones. “It’s like point, shoot, return,” Panay continued. “I think where we’re headed is a world [that] starts with ambient devices.” – https://techcrunch.com/2026/10/15/amazon-devices-chief-panos-panay-sees-a-future-where-ai-devices-replace-the-need-to-use-apps/

When it came to Cal AI’s viral success, a small army of influencers paid off – Zach Yadegari has enjoyed an immense amount of success for someone who only recently graduated from high school. Founder of the AI-fueled calorie tracking app Cal AI, Yadegari is only 19 but his company has rocketed to viral fame over the past two years. At TechCrunch Disrupt on Thursday, Yadegari shared details of that surprising rise to prominence. Cal AI, which was founded in 2024, exploded onto the internet thanks largely to Yadegari’s strategy of deploying a small army of influencers to promote the brand. The company generated over $30 million in revenue in two years and was acquired by MyFitnessPal for an undisclosed amount of money in March. Discussing his social strategy onstage, Yadegari noted that an early sign of success came when an influencer’s post about them led to approximately a thousand new users over the course of two days. – https://techcrunch.com/2026/10/15/when-it-came-to-cal-ais-viral-success-a-small-army-of-influencers-paid-off/

Tesla reveals SpaceX-powered Roadster that flies and falls with style – Tesla just revealed its redesigned second-generation Roadster with SpaceX thrusters, a vehicle CEO Elon Musk has called “the best of the last of the human-driven cars”. Musk showed off the car at an event outside Waco, Texas, with a demonstration that saw the car use a mixture of high-powered fans and thrusters to drive up an inverted ramp, spin around, and ease itself to a landing on the pavement below. The car also used the thrusters to lift itself over a Cybercab, Cybertruck, Tesla Semi, Model Y, and Model 3 while in motion. “This will be the actual car that’s on the roads,” Musk said of the car, which Tesla showed accelerating 0 to 60 miles per hour in one second. “It literally is just going to puff itself up, and uh, and be able to do this, these maneuvers whenever you’d like.” – https://techcrunch.com/2026/10/15/tesla-reveals-spacex-powered-roadster-that-flies-and-falls-with-style/