Vort3x | Cybersalon | September 15, 2026
Prepared by Wendy M. Grossman.
Vort3x, published on the 15th of each month, aims to pick out significant developments in the intersection of computers, freedom, privacy, and security for friends near and far. The views expressed in these stories do not necessarily reflect those of Cybersalon, either individually or collectively.
Contents: Cybersalon events | News | Features | Diary
Cybersalon Events
Day Techno-Progressive Futures (London)
19 and 20th September
11 am 19th – Eva Pascoe on “Future of Work & Commerce”
Future of AI
5th December, London 12pm
Pause AI rally in Central London – volunteers needed
Accountability, Preventability, Transparency of AI is not an option
Past Events
V&A Museum: 3D Glitch Art, Re-claiming Public Space for Political Protests
Notes from V&A panel with Julieta Gil and Pita Arreola (arebyte Gallery) 5/9/2026
NEWS
London Data Center Will Emit Carbon of a Million Households
———————————————————————-
Planning documents for a hyperscale data center in North Ockendon, in outer London, show that the data center will produce emissions equivalent to 8,480 flights to New York or a year’s power for 1 million UK households a year and 72 tonnes of carbon dioxide equivalent over its lifetime, Matthew Pearce reports at the Guardian. If approved, the data center will use 218 hectares of greenbelt and be one of the largest data centers in Europe. Overall, the UK’s grid operator, Neso, has warned, data centers have requested 72.8GW of connections by 2039, nearly 60% more than the UK’s peak power demand in 2025. In the planning application, developer Digital Reef describes the project as a “unique opportunity to create a sustainable data center campus of the future”.
Google escapes break-up
———————————————————————-
Judge Leonie Brinkema of the Eastern District of Virginia, who in 2025 ruled that Google broke the law by illegally monopolizing the publisher advertising market, has declined the US Department of Justice’s request that she order Google to divest the AdX exchange, but will instead, allow the company to adjust its business practices, Lauren Feiner reports at The Verge. The details of the ruling have not yet been published.
Comment: Google was ruled a law-breaking monopolist in three separate trials in the last three years. Brinkema’s was the third. In the second, Judge Amit Mehta declined to break up its search business or order the divestiture of Chrome; appeals by both Google and the DoJ are still proceeding. In the first, brought by Epic Games, the parties reached an agreement in March under which Google would redesign its app store practices. In other words, Google has so far escaped all three adverse judgments intact. Brinkema is reported to have said before the ruling that she wanted avoid years of appeals if she ordered a breakup, and that behavioral modifications are easier to enforce.
Dark Web Service Offers Millions of Scans Collected for Identity Verification
———————————————————————-
A service collected more than 153 million scans of drivers’ licenses from people in the US and Canada by regularly exfiltrating data, Brian Krebs reports at Krebs on Security. Many of the licenses were linked to renting a car; others were traced to legal marijuana dispensaries. The latter led Krebs to connect the breach to the identity verification provider idscan.net, which is now investigating. Soon after publication, a service selling the ID scans vanished from the dark web.
Comment: As many have commented, this case lends weight to the concern that age verification will expose all of us to much greater risk by creating giant pools of identity information.
AI Nearly Derails Train Carrying Hazardous Materials
———————————————————————-
A dispatcher working for the US railroad company BNSF averted a “public catastrophe” in eastern Washington when a railway worker undid a command issued by an AI-powered dispatching system that directed a 146-car freight train carrying hazardous materials onto the wrong track, Eric Rosane reports at TriCity Herald. BNSF took the system offline shortly afterwards, but reenabled it a few weeks later, according to the Brotherhood of Railroad Signalmen (a union) in a letter sent to the Federal Railroad Administration. BNSF has turned the system off again following a request from the FRA, and the union is asking for the system to remain off permanently.
Comment: The sheer size of the train is scary enough; the story shows how AI is penetrating into all sorts of places you wouldn’t think of, filling our lives with unknown risks.
Texas Court Suspends Luxembourg-Owned Domain Name Worldwide
———————————————————————
A Texas county court has ordered Verisign, the registry for the .com top-level domain, to lock the domain of a pornography site belonging to Luxembourg-based Kick Online Entertainment because it failed to comply with Texas’s age verification law, Larrison Manygoats reports at Texas Insider. Locking the domain makes it unavailable worldwide. In 2025, the US Supreme Court rejected a First Amendment challenge to the Texas age verification law. At Reason, Matthew Petti reports that on August 26 the US Department of Justice has placed the Italian hacktivist group Austici/Inventati, which provides secure email and web hosting to about 20,000 activist groups worldwide, on a terrorism watchlist. The designation immediately ends A/I’s ability to collect payment through any US-linked bank, credit card, or payment service, and ordered the Public Internet Registry, which runs the .org top-level domain, to suspend A/I’s austici.org domain. On August 28, A/I announced it was shutting down. On September 6, the Italian branch of the Internet Society published a letter to PIR asking it as the steward of .org, a supposedly safe place for non-profit organizations, to ensure that no single government can order it to take such an action in future.
Comment: The pole position occupied by the US as the home of such central Internet organizations as PIR, Verisign, and the Internet Corporation for Assigned Names and Numbers has long been a concern. These two cases show why.
FEATURES & ANALYSIS
OpenAI Agents Clustered to Attack Hugging Face
————————————————————–
In this article, Ann-Marie Corvin summarizes two new papers from METR and Redwood Research that provide more details about OpenAI agents’ attack on Hugging Face. OpenAI has said the attack was largely driven by reward hacking. The papers find that peer behavior was a crucial par, and that the attack was mounted by hundreds of agents that communicated with each other via a tool meant for downloading software to ignore ethics and cheat to complete seemingly impossible tasks. OpenAI says it is now training agents to be more skeptical of the instructions issued by other agents, and also to stop or ask for clarification when presented with tasks that seem broken or impossible. In a second article, Corvin discusses a study that found that compromised AI agents can recruit others, and then they do the cluster develops a shared language.
Comment: Security experts continue to become more alarmed as more details about the attack on Hugging Face continue to emerge.
French State-Backed Team Used Malware Downloaded from Github to Crack Encrochat Phone Network
————————————————————–
In this article at Computer Weekly, Duncan Campbell and Bill Goodwin reveal the method the secret French hacking group used in 2020 to break into tens of thousands of phones secured by the encryption software Encrochat. Czech researchers have found that the French group used an unpatched security vulnerability in Android first identified in 2017 and malware downloaded from Github. The discovery may restart legal cases in Britain’s Investigatory Powers Tribunal, which has been adjourned for more than two years while awaiting information on how the hack was carried out and, according to lawyers who say they’ve been waiting for this information for six years, important in trials and appeals in Europe.
Spy Device Detectors Offer False Sense of Safety
——————————————————————–
In this article at the Guardian, Priya Bharadia, citing a study from University College London’s computer science department, warns that detectors claiming to detect spy devices may give victims of domestic violence a false sense of safety. Lead author Akhil Polamarasetty says the study found wide disparities in capability among the more than 1,000 covert surveillance products, including shower heads and plugs with built-in cameras, and 500 detection tools they identified and that are available from mainstream online retailers. Detection tools were often not discreet enough, potentially warning abusers of their presence. Also at the Guardian, Jordyn Beazley reports that in Australia Enrico Pucci has been convicted of 30 offenses of domestic violence against his former wife; among them, he took control of her Tesla, setting up a parental control profile that allowed him to control the interior temperature, speed, locks, and curfew hours. At the Plutopia News Network podcast, the leader of the UCL research group, Leonie Tanczer, talks about her years of work studying these issues.
In Search of Digital Sovereignty
———————————————————————-
In this excerpt at Rest of World from her new book, Rulers: Corporate Power in the Age of AI and the Cloud, Cecilia Rikap compares the approaches to digital sovereignty taken by the European Commission, China, Brazil, and India. China is the only one that seems able to decouple from US technology. At the Guardian, in an extract from another new book, The Rise and Fall of the Artificial State, Jill Lepore puts the threat the technology sector now poses to democracy into the historical context of the growth of automation from the 1950s onward, charting change in news, politics, and the rise of techno-libertarians.
New Designs for Freenet
——————————————————————–
In this YouTube video, Ian Clarke, who in the 1990s created Freenet as a censorship-proof Internet service, describes his most recent work on it as an antidote to increasing centralization and government regulation of the Internet. Freenet as he now conceives it is “services without servers”: an array of peer-to-peer nodes that can run apps enabling familiar services like live chat, blogging, and software repositories while ensuring that private data always remains stored on your own system. In his design, instead of using a remote app on a server that stores your data, like Gmail, you download the software and it operates on your stored data on your own system. Rules are enforced by open-source code. His point is not that Freenet and its apps will replace familiar sites overnight, but to show that decentralization and interoperability are possible.
Comment: In 1999 Freenet terrified media commentators even though it had several problems: it was very slow, it was difficult to find anything, and participation meant hosting unknown material on your own system, a risk many preferred not to take. Decades on, the design is much more sophisticated and focused on full decentralization. With governments, courts, and companies increasingly inclined to impose the will of a single organization on the worldwide Internet through pressure on its central points of failure, new designs like these could offer important methods for checking abuse of power.
DIARY
—————————————-
October 6-7, 2026
Online
Most of FediForum will be an unconference. That means that you, the attendee, will help determine what subjects are discussed. So bring your session ideas! You can mention some during registration, but you can change your mind until the very morning of FediForum when we together make the agenda. We will also again see speed demos of cool Open Social Web software using open protocols such as ActivityPub and ATProtocol.
—————————————-
October 7, 2026
Belvol, Luxembourg
OSCL aims to create a vibrant and inclusive platform where public sector organizations, innovators, business leaders, humanitarian/non-governmental organizations, and enthusiasts can connect, share knowledge, and collaborate on the future of open source technologies. OSCL is a balanced conference where technical and non-technical people can meet to discuss the real-world impact of open source, from code and infrastructure to community, businesses and institutions.
—————————————-
October 16, 2026
Brussels, Belgium, and online
Organized by European Digital Rights (EDRi), in collaboration with its partners the Research Group on Law, Science, Technology & Society (LSTS) at Vrije Universiteit Brussel (VUB), Privacy Salon vzw , the Institute for European Studies (IEE) at Université Saint-Louis – Bruxelles, and the Racism and Technology Center, Privacy Camp brings together digital rights advocates, activists as well as academics and policy-makers from all around Europe and beyond to discuss the most pressing issues facing human rights online.
—————————————-
October 28-30, 2026
Barcelona, Spain
Mozilla Festival, aka MozFest, is three days of bold conversations, hands-on building, and ideas worth traveling for. This year’s event is partnering with the City of Barcelona and taking part in Open Tech Week — a city-wide celebration of the people shaping a better digital world.
—————————————-
November 4-6, 2026
Washington, DC, USA
The Privacy + Security Forum brings together the most seasoned thought leaders in the areas of privacy and security law for rigorous deep-dive sessions that deliver hands-on activities and practical takeaways for conference participants.
—————————————-
November 9-12, 2026
Lisbon, Portugal
Web Summit runs the world’s largest technology events, connecting people and ideas that change the world. Web Summit’s mission has been to create software that enables meaningful connections between the CEOs, founders, investors, media, politicians and cultural figureheads who are reshaping the world
—————————————-
November 13-14, 2026
Bolzano, Italy
The South Tyrol Free Software Conference, SFSCON, is one of Europe’s most established annual conferences on Free Software. SFSCON promotes the use of Free Software in digital infrastructures as a tool to achieve greater innovation and competitiveness. Here decision-makers and developers meet, learn and get inspired.
—————————————-
December 14-18, 2026
Nairobi, Kenya
The 21st annual meeting of the Internet Governance Forum, which serves to bring people together from various stakeholder groups in discussions on digital public policy, is expected to draw more htan 9,000 participants from 165 countries to contribute their perspectives on the most pressing issues, priorities, and challenges in the governance of digital technologies.
—————————————-
December 27-30, 2026
Hamburg, Germany
Over the last 40 years, the Chaos Communication Congress has become a Europe-wide renowned event with more than 17,000 participants annually, drawing an ever-growing group of international guests.
—————————————-
February 9-10, 2027
London, UK
The 2027 State of Open Con will focus on Open Technology including open-source software, open hardware, open data, open innovation, open standards, and the value that the open community brings to the UK and its digital economy. Alongside a diverse range of topics, the event promises to include a diverse range of speakers and participants.
Computers, Privacy, and Data Protection
—————————————-
March 8-11, 2027
CPDP offers the cutting edge in legal, regulatory, academic and technological development in privacy and data protection, gathering cademics, lawyers, practitioners, policy-makers, industry and civil society from all over the world. Held annually in Brussels, CPDP offers an arena to exchange ideas and discuss the latest emerging issues and trends. This unique multidisciplinary formula has served to make CPDP one of the leading data protection and privacy conferences in Europe and around the world.
—————————————-
April 26-30, 2027
Venice, Italy
The Venice Privacy Symposium aims to promote international dialogue, cooperation and knowledge sharing on data regulation, compliance, and innovative technologies and to present top level experts shedding light on the latest developments in data governance, regulatory compliance and innovative technologies.
