Is Elon Musk using SpaceX to kneecap Starlink access in Ukraine on behalf of President Donald Trump? A Ukrainian military official claimed Wednesday that Musk’s SpaceX had interrupted functioning of the country’s military-grade Starlink terminals, Reuters reported. Presidential adviser Serhiy Beskrestnov wrote on Telegram that Ukrainian defense forces “did not receive warning about this limitation, and this led to the disruption of several missions.” The disruption comes just days after Trump publicly criticized Ukraine for continuing to attack Russian oil refineries in the wake of the U.S. lifting sanctions on the sale of Russian oil, and called for the removal of  Ukrainian President Volodymyr Zelenskiy. Musk weighed in on Trump’s comments on social media Saturday, writing: “Zelensky has gone too far.” Polish Foreign Minister Radoslaw Sikorski, whose country pays for Ukraine’s Starlink access, responded to reports of the disruption. “Hey, @elonmusk, big man, are you playing politics with your products again? As your European customers, we are watching you,” he wrote on X. Musk did not respond. If the disruption really is some kind of political intervention, this is only further proof that Europeans—or really, anyone in the world—cannot rely on any of Musk’s products, as they will be at the whims of the billionaire technocrat that still finds himself at Trump’s beck and call. In a last-ditch attempt to alleviate financial strain ahead of the midterm elections, the Trump administration announced last week that the U.S. would temporarily waive sanctions on Russian diesel sold to the global market through April 2027. Trump claimed that Russia would “immediately” supply more than roughly 2.2 million barrels of diesel fuel to the American and global marketplace. (A small drop in the bucket, all things considered.)

Intended audience: non-technical citizens of the web

Three of the biggest data centres in Russia were taken offline this weekend - dramatically changing the social media landscape.

Recently, multiple organisations have disclosed how clusters of so-called “rogue” AI agents attempted to break into websites and online services, sometimes successfully. Agents from OpenAI’s environment, in particular, are known to have used other public wikis (collaboratively edited websites not owned by us) to communicate and coordinate with each other. These types of successful intrusions can expose sensitive data or disrupt website services that users rely on, while clusters of agents can attempt attacks at a scale that is difficult for defenders to manage. They affect people behind the websites who may not understand the nature of the attack, or have the tools to effectively fight back. For a site like Wikipedia, agents might find and use security vulnerabilities or make misleading edits at scale. Wikipedia’s volunteer editors and the Wikimedia Foundation’s security teams have to detect and undo that activity. The Wikimedia Foundation conducted its own investigation to see whether Wikimedia websites had been similarly affected by AI agents, focusing on those operated by OpenAI. We can confirm that we have discovered some activity by these “rogue” OpenAI agents on Wikimedia platforms. The unauthorized bot activities included edits to our wikis, some unsuccessful attempts to exploit a public note-taking tool we host, and heavy traffic, which are described more below. We did not find any evidence that our systems were used for coordination among agents, nor did we find any evidence of our systems or data being compromised. However, we are concerned about what could have occurred here, the difficulty and effort involved in investigating and attributing this activity, and the growing risks of agentic AI activity on our platforms in general. The open web is a public good. We should not allow this behavior to become the “new normal” for the people or organizations that maintain it. In summary, we saw: * **Wiki editing:** We’ve identified edits to Wikimedia wikis that we believe are from AI agents operated by OpenAI. These edits were not published to pages with visibility to general readers; almost all of them were testing edits in “sandbox” areas of the wiki. It also included a few edits to the configuration for a citation tool, which we believe were potentially malicious edits that were intended to misuse this tool as a proxy for fetching data from remote services. While Wikipedia policies allow bots to edit when they are disclosed and approved by the community, none of those approvals were sought in these incidents. * **Etherpad probing and use:** Agents we believe to be operated by OpenAI made some unsuccessful attempts to compromise our public Etherpad, a note-taking tool we host as a community service. Agents unsuccessfully tried to use it to fetch data from other websites as a proxy. Other agents also likely operated by OpenAI took notes about their tasks, though this did not appear to turn into coordination. * **Excessive data downloading:** Agents we believe to be operated by OpenAI made millions of automated requests to our public APIs to access the knowledge on Wikimedia projects, crawled millions of pages (mainly from our projects Wikidata and Wikimedia Commons), and made hundreds of thousands of data queries to the Wikidata Query Service (WQDS). This traffic may have contributed to a partial outage on WQDS in May. As a non-profit technology host of some of the largest and most widely used open knowledge platforms in the world, we are deeply concerned about the impact of “rogue” AI agents on platforms like ours, which are built by volunteers from around the world and rely on the promise of the open internet. Incidents like this one, and the many others that have been (and are still being) uncovered, illustrate how AI agents can drain resources and crash servers, as well as attempt to compromise trustworthy information. Over the past 25 years, Wikipedia has grown into one of the most popular and trusted websites in the world, with more than **67 million articles across over 300 languages, and up to 15 billion page views per month**. Through an open, transparent, and collaborative process, volunteers work to ensure that knowledge remains neutral, reliable, and accessible to everyone. Wikipedia is one of the highest-quality datasets used in training Large Language Models (LLMs), and its knowledge forms the backbone of information on the internet, powering AI chatbots, search engines, voice assistants, and more. Wikipedia was designed for humans – and agentic behavior clearly poses challenges that no one has solutions for. Because of our unique and successful knowledge creation model, Wikimedia’s volunteers are the ones who come in first contact with, and clean up the mess left behind by AI agents. Rising bot traffic and agentic activity is showing a real impact on the Wikimedia projects and the infrastructure that makes it available for millions of users globally. In 2025, the Foundation reported that its bandwidth usage had increased by 50% due to the surge of bot activity on its websites since 2024. At the same time, 65% of the most resource-consuming traffic on its projects was coming from bots. This intense pressure on our infrastructure not only adds costs for servers and humans, but if left unaddressed, can block human visitors by overloading systems and causing outages. We are already paying for costs that come with the increased activity. Wikimedia’s volunteers have stayed resilient so far in tackling emerging challenges on our platforms, but we also want to say: it doesn’t need to be this way. While OpenAI admits to agents behaving “unpredictably”, they must also acknowledge their responsibility to monitor and prevent these risks. AI companies are not doing enough to secure their systems and protect the public from the harm they cause. That burden is falling onto everyone else, including smaller organizations. At a minimum, their systems should operate in a way that non-profit website owners like us can easily identify, and choose how they interact with our services. The web enables so much: to connect with friends and family, to register for school, to plan a trip across town, to buy groceries, and to learn about the world from Wikipedia. Bots and agents are part of the future of the web, and the companies who unleash and profit from them must directly help avoid and repair damage they can do. Our collective priority should be the health of the overall web ecosystem so that it continues to benefit all people – not just a handful of billionaires. Wikimedia plays a critical role in stewarding the knowledge commons, but we cannot do it alone. We invite everyone who is building the future of the web to join us in protecting the open, shared resources that make that future possible.

De Nederlandse overheid brengt de DigiD-app uit voor pc's, wat eerder al was aangekondigd. De pc-app is bedoeld voor mensen die geen smartphone hebben. In Nederland zijn dat er een miljoen, aldus de overheid. De app draait op Windows 11 en Mac-computers met macOS 15 (Sequoia) of hoger.

Tot nu toe legde het Witte Huis al sancties op aan een aantal rechters en aanklagers, maar werd nog geen actie ondernomen tegen het Strafhof zelf.

Canada’s counter-tariffs range from 15 to 50 per cent.

### Crowdfunding for this project is now open Rather than a full crowdfunding setup, this time finance will be organised via my newsletter subscription system. There is a special subscription newsletter for the duration of the project (€5/month for 3 months). You can also give a small tip for my work on this project (suggested €10). If you would like to support my work more permanently, consider signing up to my weekly railway newsletter. --- ### What is the idea? In August 2026 a new cross border passenger rail connection will open between Tornio in Finland and Haparanda in Sweden. This then allows an uninterrupted rail journey from Kemijärvi in Finnish Lapland to Lagos in Portugal, the two places furthest apart in the EU that can be connected by train. I, Jon Worth, the person who worked this out, aim to be the first person to complete the whole trip, and set the first record time for this journey. ### Why would someone possibly want to do this? Because it’s fun! What’s not to like about a massive and crazy journey? Seriously though, telling stories through concrete experiences is a really powerful way to explain complicated political and technical railway topics in a way that is fun and relatable. The #CrossBorderRail project as a whole has shown this, and this record attempt will build on that work. Just even working out the possible timetable for this trip has caused so many headaches so as to make the whole project worthwhile. Stuck in Boden for 5 hours, why? And when doing the trip there will be further positive and negative experiences aplenty. Locomotives in Boden station – a view I can admire for 5 hours on the trip And hell, if I don’t do it, some rail Youtuber will likely do it for clickbait instead. If I do it then I can manage to get some solid political analysis out of it too. ### What is the definition of the longest train journey? The two places in the EU that are the greatest geodesic or great-circle distance (i.e. how the crow flies) distance apart that can be connected by multiple trains. Start and end points are Kemijärvi, Finland (station geo-coordinates: 66°43′25″N 027°24′14″E or 66.72383, 27.40368) and Lagos, Portugal (station geo-coordinates: 37°6′31″N 8°40′18″W or 37.1085361111, -8.6716722222). The geodesic distance between them can be calculated using this tool, and it is **4008.33km**. The route-km is **6165.8km** , although that does not really concern us here. ### What are the rules of the record attempt? A request to Guinness World Records was submitted in spring to 2026 to adjudicate on the criteria, but as their processes are notoriously slow and I am still waiting for a reply – so I am going to have to set my own criteria. These are: 1. For the record attempt only regular, public train services can be used – trains the general public can use. 2. Train services on tracks of the Single European Railway Area (SERA) can be used (this is the best definition of an EU rail system there is), and this means non-EU countries can be transited if necessary if their tracks are part of the Area. 3. Other public transport, including on tracks, is not allowed – so that means no trams, metros or suburban trains that are not part of the Single European Railway Area. 4. When changing trains, stations are simply defined as anything within the same building – walking out of stations and onto public highways is not permitted. 5. Exiting stations for whatever purpose (to purchase food and refreshments, for meetings, for hotel stays) is allowed, providing the next leg starts at the station where the previous one ended. De facto these rules mean that transiting Switzerland would be permissible, using the Metro or RER to cross Paris is not permitted (as it is a standalone network not included in the SERA), and walking from Madrid Puerta de Atocha to Atocha–Cercanias is allowed as these are in the same building. High speed trains at Madrid Atocha To make a competition of this there is a time element: the record is the **fastest trip on the longest train journey in the EU**. This is to be timed from the moment the doors close on the train departing Kemijärvi and the moment the doors open at Lagos, or vice versa if the journey is attempted in reverse. ### Can other people attempt this? Of course! I see this as something like the Tube Challenge, and as better services, improved timetables and elimination of temporary speed restrictions happen in subsequent years it ought to be possible to eventually complete this trip more quickly than in 2026. I would also like more people to be part of my record attempt – please see “How can you help?” below. All I would ask is if you are trying it yourself you let me know, or link to this post. And if you complete it, tell me your total time. Contact me via this form. ### How long will the trip to take? My current detailed timetable (below), and attempting the trip in the Kemijärvi – Lagos direction, has a total trip time of 100 hours and 9 minutes (so just over 4 days). Of this 61 hours and 8 minutes are on trains. The average speed of the trains themselves is 102.46km/h, the average speed of the trip inc. time not moving is 61.52km/h. ### When will it start? Provisional start time is 19:00 EEST on 7th September 2026 at Kemijärvi station. Due to other commitments I cannot attempt it in August immediately once Tornio – Haparanda has opened. ### How much will the trip cost? At a later stage I will calculate the total theoretical ticket price using regular tickets, but I am going to make the trip using an Interrail pass. This will require a 5 day pass, and with reservation costs on top, should be possible for between €400 and €500. ### What happens if something goes wrong? The timetable has been planned with what I estimate are adequate buffers to accommodate normal delays throughout the trip. If something more serious goes wrong I will have to re-plan and re-book accordingly, even if this then means arriving in Lagos a day later. ### Will there be merchandise? Absolutely! This will be ready shortly! And there will postcards too! ### How can people help? Above all I want to meet interesting people en route – especially in Kemijärvi and Lagos, the start and end points, and at important points en route, notably Tornio-Haparanda, Boden, Stockholm, Hamburg, Strasbourg, Lyon and Madrid. And people are welcome to join me on board for a leg, or part of a leg. You can find the full provisional timetable below. Rather than a full crowdfunding setup, this time finance will be organised via my newsletter subscription system. There is a special subscription newsletter for the duration of the project (€5/month for 3 months). You can also give a small tip for my work on this project (suggested €10). If you would like to support my work more permanently, consider signing up to my weekly railway newsletter. The whole idea is crazy and compelling enough to mean someone might want to cover it in the media or come along and film it all. I am open to any sort of partnership of that sort too! There will be all sorts of ways to follow the project online, and plenty of data, photos and video generated. For the moment please use this form to contact me about any aspect of the project. ## The detailed nerdy stuff! ### What’s the detailed timetable? In mid July it looked like something was changing in northern Sweden – on 10th August train times Haparanda – Boden were changed to better match the Finland – Sweden cross border trains, but it seems this better timetable only applies on the launch day. So, for the moment, this looks to be the timetable for the record attempt – with 5 hours stuck in Boden. Mon 07/09| 1| 274| VR| 19:00| Kemijärvi| Kemi| 22:27 ---|---|---|---|---|---|---|--- Night in Kemi, hotel| | | | | | | Tue 08/09| 2| IC 401| VR| 06:08| Kemi| Haparanda| 05:45 | 3| R 7183| Norrtåg| 07:40| Haparanda| Boden Centralstation| 09:08 | 4| R 7107| Norrtåg| 14:05| Boden Centralstation| Umeå Central| 17:24 | 5| D 91| SJ| 20:19| Umeå Central| | Night in night train| | | | | | | Wed 09/09| | | | | | Stockholm Central| 06:18 | 6| X2 523| SJ| 07:22| Stockholm Central| Lund C| 11:40 | 7| R 1065| Øresundståg| 11:51| Lund C| København H| 12:44 | 8| ECE 397| DSB, DB Fernverkehr| 14:22| København H| Hamburg Hbf| 19:01 | 9| NJ 471| ÖBB| 22:11| Hamburg Hbf| | Night in night train| | | | | | | Thu 10/09| | | | | | Offenburg| 06:39 | 10| RS4 (87408)| SWEG or SNCF Voyageurs for TER Fluo| 07:05| Offenburg| Strasbourg| 07:34 | 11| TGV 9877| SNCF Voyageurs| 09:04| Strasbourg| Lyon Part Dieu| 12:59 | 12| AVE 9742| Renfe| 14:35| Lyon Part Dieu| Barcelona Sants| 19:23 | 13| IRY 6200| Iryo| 19:40| Barcelona Sants| Madrid Atocha| 22:42 Night in Madrid, hotel| | | | | | | Fri 11/09| 14| A 190| Renfe| 08:51| Madrid Atocha Cercanias| Badajoz| 13:26 | 15| R 482| CP| 14:09| Badajoz| Entroncamento| 15:55 | 16| IC 512| CP| 15:59| Entroncamento| Lisboa Oriente| 16:52 | 17| IC 184| CP| 17:23| Lisboa Oriente| Tunes| 19:53 | 18| R 5914| CP| 20:07| Tunes| Lagos| 21:09 Some of these connections are quite tight, although in most (but not all!) cases I have some backups ready. I will re-plan as necessary during the journey. ### Is this mapped? The zoomable version of the map is here. ### Why is this using a geodesic distance definition? There are two other potential definitions, but each has its downsides. The first – the most route-km on a single train (currently the night train Berlin Hbf – Stockholm Central, 1444.1km) is simple to calculate, but anyone who has taken that train holds the record. There is no skill, no challenge, in doing that one. A common critique of the geodesic definition is “but aren’t there longer routes?” and yes, you could keep adding more and more via stations to add route-km, but more km for the sake of it doesn’t really strike me as a useful pursuit. However you could just about come up with a definition along the lines of “the most route-km on multiple trains in the EU where no shorter route is available”. But trying to work out the route of this one was so ridiculously complicated that I think it ought to be eliminated. I _think_ it is Savonlinna, Finland to Vila Real de Santo António, Portugal, and the total is 6744.5km, but I am not at all confident about this due to the complexity of the calculation. This is all explored in more detail here. So one train is too simple, and most route-km on multiple trains in the EU where no shorter route is available is too complicated. Geodesic distance on multiple trains is a good compromise. ### What is the route to Kemijärvi, and back from Lagos? By train of course! There are a few borders I am missing in the Baltics, and at the Portugal – Spain border, for my main #CrossBorderRail work, and I will check out those places before and after the record attempt. You can find all my route planning for all of that in a spreadsheet in Nextcloud here, and all of this is mapped here. I aim to get to 20 new rail borders en route to Kemijärvi and back from Lagos, and after Innotrans: Eglaine 🇱🇻 – Czarny Brod 🇱🇹 (Possible former 600mm gauge line during WWI) Eglaine 🇱🇻 – Obeliai 🇱🇹 Subate 🇱🇻 – Obeliai 🇱🇹 (600mm gauge) Akniste 🇱🇻 – Rokiškis 🇱🇹 (Former 600mm gauge line during WWI) Nereta 🇱🇻 – Suvainiškis 🇱🇹 (Former 600mm gauge line during WWI) Rucava 🇱🇻 – Šventoji 🇱🇹 (Former 600mm, later 750 mm) Priekule 🇱🇻 – Skuodas 🇱🇹 Vainode 🇱🇻 – Luše 🇱🇹 Renge 🇱🇻 – Mažeikiai 🇱🇹 Mõniste 🇪🇪 – Ape 🇱🇻 (Former 750mm gauge line) Valka 🇱🇻 – Mõniste 🇪🇪 (Former 750mm gauge line) Möisaküla 🇪🇪 – Ipiki 🇱🇻 Barca de Alva 🇵🇹 – Fregeneda 🇪🇸 Marvao Beira 🇵🇹 – Valencia do Alcántara 🇪🇸 La Farga de Moles 🇪🇸 – Sant Julià de Lòria 🇦🇩 (Future tramway) Szöreg 🇭🇺 – Novi Kneževac 🇷🇸 Szöreg 🇭🇺 – Banatsko Aranđelovo 🇷🇸 Cenad 🇷🇴 – Apátfalva 🇭🇺 Ściborzyce Małe 🇵🇱 – Osoblaha 🇨🇿 Pilszcz 🇵🇱 – Opava západ 🇨🇿