
Chaos Computer Club - recent events feed
4,903 episodes — Page 42 of 99
When Vibe Scammers Met Vibe Hackers: Pwning PhaaS with Their Own Weapons (39c3)
When Vibe Scammers Met Vibe Hackers: Pwning PhaaS with Their Own Weapons (39c3)
What happens when AI-powered criminals meet AI-powered hunters? A technical arms race where both sides are vibing their way through exploitation—and the backdoors write themselves. In October 2025, we investigated Taiwan's fake delivery scam ecosystem targeting convenience store customers. What started as social engineering on social media became a deep dive into two distinct fraud platforms—both bearing the unmistakable fingerprints of AI-generated code. Their developers left more than just bugs: authentication flaws, file management oversights, and database implementations that screamed "I asked LLM and deployed without reading." We turned their sloppiness into weaponized OSINT. Through strategic reconnaissance, careful database analysis, and meticulous operational security, we achieved complete system access on multiple fraud infrastructures. By analyzing server artifacts and certificate patterns, we mapped 100+ active domains and extracted evidence linking thousands of victim transactions worth millions of euros in fraud. But here's the twist: we used the same AI tools they did, just with better prompts. The takeaway isn't just about hunting scammers—it's about the collapse of the skill gap in both offense and defense. When vibe coding meets vibe hacking, the underground economy democratizes in ways we never anticipated. We'll share our methodology for fingerprinting AI-assisted crime infrastructure, discuss the ethical boundaries of counter-operations, and demonstrate how to build sustainable threat intelligence pipelines when your adversary can redeploy in 5 minutes. This talk proves that in 2025, the real exploit isn't zero-day—it's zero-understanding. Our journey began with a simple question: why are so many people losing money to fake convenience store delivery websites? The answer led us through two distinct criminal architectures, both exhibiting characteristics of large language model–assisted development. Case 1 ran on PHP with backup artifacts exposing implementation details and query manipulation opportunities. The installation package itself contained pre-existing access mechanisms—whether this was developer insurance or criminal-on-criminal sabotage remains unclear. We leveraged initial access to bypass security restrictions using protocol-level manipulation and extracted gigabytes of operational data. Case 2 featured authentication bypass vulnerabilities that granted direct administrative access. The backend structure revealed copy-pasted code patterns without proper security implementation. Throughout both systems, we observed telltale signs of AI-generated code: verbose documentation in unexpected languages, inconsistent coding patterns, textbook-like naming conventions, and theoretical security implementations. Even the UI revealed LLM fingerprints—overly polished component layouts, placeholder text patterns, and design choices that felt distinctly "tutorial-like." These weren't experienced developers—they were operators deploying what LLMs gave them without understanding the internals. The irony? We used AI extensively too: for data parsing, pattern recognition, attack surface mapping, and intelligence queries. The difference was intentionality—we understood what the output meant. Using open-source intelligence platforms and carefully crafted fingerprints, we mapped over a hundred active domains following similar patterns. Each one shared the same architecture, the same weaknesses, the same developer mistakes. This repeatability became our advantage. When scammers can redeploy infrastructure in days, you don't attack individual sites—you automate the entire reconnaissance-to-evidence pipeline. This talk demonstrates practical techniques for mass-scale fraud infrastructure fingerprinting, operational security considerations when investigating active criminal operations, and methods to recognize AI-generated code patterns that reveal threat actor sophistication. We'll discuss the ethical boundaries of counter-fraud operations and evidence preservation for law enforcement, along with automation strategies for sustainable threat intelligence when adversaries rebuild faster than you can report. The demonstration will show how to go from a single suspicious domain to a network map of 100+ sites and thousands of victim records—using tools available to any researcher. This isn't a story about elite hackers versus criminal masterminds. It's about two groups equally reliant on AI vibing their way through technical problems—one for fraud, one for justice. The skill barrier has collapsed. The question now is: who has better context, better ethics, and better coffee? Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/when-vibe-scammers-met-vibe-hackers-pwning-phaas-with-their-own-weapons
51 Ways to Spell the Image Giraffe: (39c3)
Generative AI models don't operate on human languages – they speak in **tokens**. Tokens are computational fragments that deconstruct language into subword units, stored in large dictionaries. These tokens encode not only language but also political ideologies, corporate interests, and cultural biases even before model training begins. Social media handles like *realdonaldtrump*, brand names like *louisvuitton*, or even *!!!!!!!!!!!!!!!!* exist as single tokens, while other words remain fragmented. Through various artistic and adversarial experiments, we demonstrate that tokenization is a political act that determines what can be represented and how images become computable through language. Tokens are the fragments of words that generative models use to process language, the step that breaks text into subword units before any neural networks are involved. There are 51 ways to combine tokens to spell the word giraffe using existing vocabulary: from a single token **giraffe** to splits using multiple tokens like *gi|ra|ffe*, *gira|f|fe*, or even *g|i|r|af|fe*. In one experiment, we hijacked the prompting process and fed token combinations directly to text-to-image models. With variations like *g|iraffe* or *gir|affe* still generating recognizable results, our experiments show that the beginning and end of tokens hold particular semantic weight in forming giraffe-like images. This reveals that certain images cannot be generated through prompting alone, as the tokenization process sanitizes most combinations, suggesting that English, or any human language, is merely a subset of token languages. The talk features experiments using genetic algorithms to reverse-engineer prompts from images, respelling words in token language to change their generative outcomes, and critically examining token dictionaries to investigate edge cases where the vocabulary breaks down entirely, producing somewhat *speculative languages* that include strange words formed at the edge of chaos where English meets token (non-)sense. These experiments show that even before generation occurs, token dictionaries already encode a stochastic worldview, shaped by the statistical frequencies of their training data – dominated by popular culture, brands, platform-speak, and *non-words*. Tokenization is, therefore, a political act: it defines what can be represented and how the world becomes computationally representable. We will look at specific tokens and ask: Which models use which vocabularies? What *non-word* tokens are shared among models? And how do language models make sense of a world using a language we do not understand? Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/51-ways-to-spell-the-image-giraffe-the-hidden-politics-of-token-languages-in-generative-ai
How To Minimize Bugs in Cryptography Code (39c3)
"Don't roll your own crypto" is an often-repeated aphorism. It's good advice -- but then how does any cryptography get made? Writers of cryptography code like myself write code with bugs just like anyone else, so how do we take precautions against our own mistakes? In this talk, I will give a peek into the cryptographer's toolbox of advanced techniques to avoid bugs: targeted testing, model checking, mathematical proof assistants, information-flow analysis, and more. None of these techniques is a magic silver bullet, but they can help find flaws in reasoning about tricky corner cases in low-level code or prove that higher-level designs are sound, given a defined set of assumptions. We'll go over some examples and try to give a high-level feel for different workflows that create "high-assurance" code. Whether you know it or not, you use this type of cryptography code every day: in your browser, your messaging apps, and your favorite programming language standard libraries. Over the last 10 years or so, using mathematical proof assistants and other formal-logic tools for cryptography code has gone from a relatively new idea to standard practice. I've been lucky enough to have a front-row seat to that transformation, having started doing formal-methods research in 2015 and then switched to a focus on cryptography implementation since 2021. Code from my master's thesis project, ["fiat-crypto"](https://github.com/mit-plv/fiat-crypto), is [included](https://andres.systems/fiat-crypto-adoption.html) in every major browser as well as AWS, Cloudflare, Linux, OpenBSD, and standard crypto libraries for Go, Zig, and Rust (RustCrypto, dalek). In addition to verifying code correctness, designers of high-level protocols like Signal's recently announced post-quantum ratchet increasingly use mathematical tools (ProVerif in Signal's case) to check their work. Despite the growing popularity of these formal techniques and their relevance to personal information security, few people are aware of them, and they maintain a reputation for being hard to learn and esoteric. I'd like to demystify the topic and show examples of how anyone can use proof assistants in small, standalone ways as part of the coding or design process. My hope is that next time a colleague asks for review of a complex high-speed bit-twiddling algorithm, instead of staring at the code line-by-line, attendees of my talk will know they can write a computer-checked proof to confirm or deny that the algorithm achieves its intended result. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/how-to-minimize-bugs-in-cryptography-code
Machine Vision (39c3)
Milliarden von Kameras produzieren täglich Bilder, die zunehmend von Maschinen analysiert werden. In dieser Lecture Performance beleuchten wir die Entwicklung des maschinellen Sehens – von den frühen algorithmischen Ansätzen bis zu den heutigen Anwendungen – und schauen, wie verschiedene Künstler:innen diese Technologien nutzen und reflektieren. Anhand der beiden Arbeiten „Throwback Environment” und „Fomes Fomentarius Digitalis” betrachten wir die Nutzung des maschinellen Sehens in einem künstlerischen Feedback-Loop. Die Arbeiten machen sichtbar, was die eingesetzten Algorithmen sehen und in welchen Mustern sie operieren. Unmengen an Bilder werden Täglich in die Netzwerke hochgeladen. Doch nicht nur Menschen betrachten diese Bilder, auch Maschinen analysieren und „betrachten" sie. Wie funktioniert dieses maschinelle „Sehen" und wie wurde dieses den Computern beigebracht? Diese Lecture Performance gibt einen Überblick über die Entwicklung des maschinellen Sehens. Nach einem kurzen Einblick in die geschichtliche Entwicklung – von den ersten Ansätzen bis zu heutigen Anwendungen – betrachten wir, wie diese Technologien in unterschiedlichsten künstlerischen Arbeiten eingesetzt werden. Was reflektieren diese Arbeiten jenseits der reinen Anwendung von Machine Vision Algorithmen? Anhand der beiden Arbeiten "Throwback Environment" und "Fomes Fomentarius Digitalis" betrachten wir, wie Machine Vision in einem künstlerischen Feedbackloop genutzt worden ist und wie uns dies Perspektiven auf die Funktionsweise dieser Algorithmen eröffnet. Die Arbeiten machen sichtbar, was die Eingesetzten Alghorithmen sehen, in welchen Mustern sie operieren. Sie zeigen auch, wo ihre Grenzen liegen und was das ganze mit Baumpilzen zu tun hat. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/machine-vision-vom-algorithmus-zum-baumpilz-im-digitalen-metabolismus
Don’t look up: There are sensitive internal links in the clear on GEO satellites (39c3)
We pointed a commercial-off-the-shelf satellite dish at the sky and examined all of the geostationary satellite communications visible from our vantage point. A shockingly large amount of sensitive traffic is being broadcast unencrypted, including critical infrastructure, internal corporate and government communications, private citizens’ voice calls and SMS, and consumer Internet traffic from in-flight wifi and mobile networks. In this talk, we will cover our hardware setup, alignment techniques, our parsing code, and survey some of the surprising finds in the data. This talk will include some previously unannounced results. This data can be passively observed by anyone with a few hundred dollars of consumer-grade hardware. There are thousands of geostationary satellite transponders globally, and data from a single transponder may be visible from an area as large as 40% of the surface of the earth. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/don-t-look-up-there-are-sensitive-internal-links-in-the-clear-on-geo-satellites
Textiles 101: Fast Fiber Transform (39c3)
Textiles are everywhere, yet few of us know how they’re made. This talk aims to give you an overview over the complete transformation from fiber to finished textile. We'll be exploring fiber properties, spinning, and techniques like weaving, knitting, crochet, braiding, and knotting, followed by finishing methods such as dyeing, printing, and embroidery. You’ll learn why not only fiber but also structure matters, and how to make or hack textiles on your own without relying on fast fashion or industrial tools. Textiles play an integral part in our daily lives. If you’re reading this, chances are you’re wearing clothes or have some form of fabric within arm’s reach. Yet despite how common and essential textiles are, few of us know how they actually come to be. How do we go from a plant, animal, or synthetic polymer to a fully finished piece of clothing? This talk unravels the full transformation pipeline of textiles: starting with fibers and their properties, then spinning them into yarn, turning that yarn into textiles through weaving, knitting, crochet, braiding, knotting, and other techniques, and finally finishing them through printing, embroidery, dyeing, or bleaching. Along the way, you’ll learn why your “100% cotton” garments can feel completely different despite being made of the same fiber, how structure matters just as much as material, and what environmental impact different choices have. Whether you want to make your own textiles, hack existing ones, or finally understand why that wool sweater you washed too hot is now tiny, this talk is a crash course in most things textile, and a reminder that you don’t need industrial machinery or fast fashion to create something on your own. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/textiles-101-fast-fiber-transform
Machine Vision (39c3)
Don’t look up: There are sensitive internal links in the clear on GEO satellites (39c3)
How To Minimize Bugs in Cryptography Code (39c3)
Textiles 101: Fast Fiber Transform (39c3)
CSS Clicker Training: Making games in a "styling" language (39c3)
CSS is a programming language, and you can make games in it. Let's install NoScript and make some together! This talk is about how HTML and CSS can be used to make interactive art and games, without using any JS or server-side code. I'll explain some of the classic Cohost CSS Crimes, how I made [CSS Clicker](https://lyra.horse/css-clicker/), and what's next for the CSS scene. I hope this talk will teach and/or inspire you to make cool stuff of your own! --- *Content notes:* - Slides feature animations and visual effects - Short video clip (with music) will be played - Clicker sound at the end of the talk --- Slides will be available after the talk at: [https://lyra.horse/slides/#2025-congress](https://lyra.horse/slides/#2025-congress) Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/css-clicker-training-making-games-in-a-styling-language
Power Cycles statt Burnout – Wie Einflussnahme nicht verpufft (39c3)
Zwischen offenen Briefen, Massenmails, Petitionen und Kaffee trinken : Zwei Ex-Insiderinnen aus dem Digitalausschuss und Bundestag erzählen, wie politische Einflussnahme wirklich funktioniert. Ein ehrlicher Blick hinter die Kulissen parlamentarischer Entscheidungsfindung – mit praktischen Tipps, wie die Zivilgesellschaft ihre Energie dort einsetzt, wo sie wirklich Wirkung entfaltet. Ziel des Talks ist es, ein realistisches Bild davon zu vermitteln, wie parlamentarische Entscheidungsfindung funktioniert – und praktische Hinweise zu geben, wie man Einfluss nehmen kann, ohne dabei Ressourcen zu verschwenden. Wie bringt man politische Prozesse in Bewegung? Was passiert eigentlich mit einer Mail, wenn sie an einen Abgeordneten geht? Und wie unterscheidet sich konstruktive Interessenvertretung von übergriffigem Lobbyismus? In diesem Talk berichten Anna Kassautzki (Mitglied des Bundestags von 2021 bis 2025, stellvertretende Vorsitzende des Digitalausschusses 20. LP) und Rahel Becker (ehemalige wissenschaftliche Mitarbeiterin Digitales) aus der Innenperspektive parlamentarischer Arbeit. Chatkontrolle, Data Act, Recht auf Open Data, DSGVO, es gab viel zu verhandeln in der letzten Legislaturperiode. Anna und Rahel waren mittendrin und geben einen Einblick in die hektische - teils absurde Kommunikation mit Interessenvertretungen. Dabei liegt der Fokus immer auf der Frage: Welche Strategien braucht es, damit zivilgesellschaftliche Arbeit nicht verpufft? Zugleich geht es um die strukturellen Fragen: Wo sind die Flaschenhälse für politischen Fortschritt? Wie priorisieren Abgeordnete in einem überfüllten Kalender? Und welche Hebel kann die (digitale) Zivilgesellschaft sinnvoll nutzen, um Gehör zu finden? Denn gerade in Zeiten massiver digitalpolitischer Herausforderungen ist informierte, strategische Beteiligung notwendiger denn je. Ein Vortrag für alle, die sich in politische Prozesse einmischen wollen. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/power-cycles-statt-burnout-wie-einflussnahme-nicht-verpufft
Verlorene Domains, offene Türen (39c3)
CSS Clicker Training: Making games in a "styling" language (39c3)
Wie wir alte Flipperautomaten am Leben erhalten (39c3)
Power Cycles statt Burnout – Wie Einflussnahme nicht verpufft (39c3)
Wie wir alte Flipperautomaten am Leben erhalten (39c3)
Der Vortrag beschreibt, wie eine Gruppe von Begeisterten eine Sammlung von ca. 100 Flipperautomaten (Pinball Machines) am Leben und in spielbereitem Zustand erhält. Der Vortrag gibt einen Einblick in die verschiedenen Generationen von Flippern und deren Technik. Angefangen von elektromechanischen Geräten aus den frühen Sechzigern, über erste Prozessorsteuerungen, bis hin zu modernsten computergesteuerten Automaten mit Bussystemen. Jede Generation hat ihre technischen Eigenheiten, ihre typischen Fehlermuster und Schwachstellen. In öffentlichen Räumen sind heutzutage kaum mehr Flipper anzutreffen. Das liegt insbesondere daran, dass deren Wartung aufwändig ist, weil durch die mechanische Beanspruchung häufig Fehler auftreten. Bereits kleinste technische Probleme können den Spielspaß zunichte machen. Das Finden und Beheben von Fehlern erfordert viel Erfahrung – und manchmal Kreativität, insbesondere wenn alte Bauteile nicht mehr verfügbar sind oder kaum Dokumentation vorhanden ist. Technisch ist Sachverstand auf vielen Ebenen erforderlich, vom Schaltplanlesen über Löten und elektronische Messtechnik, bis hin zu mechanischem Know-how. Die Community der Flipper-Enthusiasten ist allerdings groß und kooperativ, sodass auch private Sammler ihre Flipper am Laufen halten können. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/wie-wir-alte-flipperautomaten-am-leben-erhalten
Verlorene Domains, offene Türen (39c3)
Was passiert, wenn staatliche Domains auslaufen - und plötzlich jemand anderes sie besitzt? In diesem Vortrag wird berichtet, wie mehrere ehemals offizielle, aber unregistrierte Domains deutscher Bundesministerien und Behörden erworben werden konnten - und welche Datenströme dadurch sichtbar wurden. Über Monate hinweg konnten so DNS-Anfragen aus Netzen des Bundes empfangen werden - ein erhebliches Sicherheitsrisiko. Unter anderem da es so möglich war Accounts zu übernehmen, Validierungen von E-Mailsignaturen zu manipulieren, Anfrage umzuleiten und im Extremfall Code auf Systemen auszuführen. (Keine sensiblen Daten werden veröffentlicht; der Fokus liegt auf Forschung, Aufklärung und verantwortungsvollem Umgang mit den Ergebnissen.) Im Rahmen der Untersuchung zeigten sich nicht nur Fehlkonfigurationen, sondern auch Phänomene wie Bitsquatting und Typoquatting innerhalb der Verwaltungsnetze. Mit dem Betrieb eines DNS-Servers und dem Erwerb von bund.ee (naher Typosquatting/Bitquatting zu bund.de) konnten u.a. zahlreiche DNS-Anfragen von Servern des Bundesministerium des Innern (BMI) und weiterer Einrichtungen des Bundes empfangen werden. Der Vortrag beleuchtet die technischen und organisatorischen Schwachstellen, die hinter solchen Vorgängen stehen - und zeigt, wie DNS-Details Einblicke in die IT-Infrastruktur des Staates ermöglichen können. Abgerundet wird das Ganze durch praktische Beispiele, Datenanalysen und Empfehlungen, wie sich ähnliche Vorfälle künftig vermeiden lassen. In anderen Ländern sind gov-Domains als TLDs längst üblich (bspw. gov.uk) - in Deutschland ist bund.de oder gov.de allerdings nicht so verbreitet wie man glaubt, unter anderem da Bundesministerien eigene Domains nutzen oder nach Regierungsbildung umbenannt werden. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/verlorene-domains-offene-turen-was-alte-behordendomains-verraten
"Passwort" - der heise security Podcast live vom 39C3 (39c3)
"Passwort" - der heise security Podcast live vom 39C3 (39c3)
Der heise security Podcast macht wieder einen Betriebsausflug nach Hamburg. Diesmal bringt Christopher seinen Co-Host Sylvester mit und spricht 90 Minuten lang über aktuelle Security-Themen vom Congress. Wir haben uns erneut einige spannende Fundstücke herausgesucht und sprechen darüber miteinander, aber auch mit unseren Gästen. Welche Themen wir besprechen ist - wie immer bei unserem Podcast - eine Überraschung. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/passwort-der-heise-security-podcast-live-vom-39c3
Recharge your batteries with us (39c3)
Prometheus: Reverse-Engineering Overwatch (39c3)
freiheit.exe - Utopien als Malware (39c3)
Trump government demands access to European police databases and biometrics (39c3)
Trump government demands access to European police databases and biometrics (39c3)
The USA is demanding from all 43 countries in the "Visa Waiver Programme" (VWP), which enables visa-free travel, to conclude an "Enhanced Border Security Partnership" (EBSP). This is intended to grant US authorities direct access to police databases in these - mostly European - countries containing fingerprints, facial images and other personal data. Anyone who refuses this forced "border partnership" faces exclusion from the visa-free travel programme. The US demand is unprecedented: even EU member states do not grant each other such extensive direct database access – normally the exchange takes place via the "hit/no-hit principle" with a subsequent request for further data. This is how it works, for example, in the Prüm Treaty among all Schengen states, which has so far covered fingerprints and DNA data and is now also being extended to facial images. The EBSP could practically affect anyone who falls under the jurisdiction of border authorities: from passport controls to deportation proceedings. Under the US autocrat Donald Trump, this is a particular problem, as his militia-like immigration authority ICE is already using data from various sources to brutally persecute migrants – direct access to police data from VWP partners could massively strengthen this surveillance apparatus. Germany alone might give access to facial images of 5.5 million people and fingerprints of a similar dimension. The USA has already tightened the Visa Waiver Programme several times, for instance in 2006 through the introduction of biometric passports and in 2008 through the ESTA pre-registration requirement. In addition, there were bilateral agreements for the exchange of fingerprints and DNA profiles – however, these may only be transmitted in individual cases involving serious crime. Existing treaties such as the EU-US Police Framework Agreement are not applicable to the "Enhanced Border Security Partnership", as it applies exclusively to law enforcement purposes. It is also questionable how the planned data transfer is supposed to be compatible with the strict data protection rules of the GDPR. The EU Commission therefore wants to negotiate a framework agreement on the EBSP that would apply to all member states. Time is running short: the US government has set VWP states a deadline of 31 December 2026. Some already agreed on a bilateral level. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/trump-government-demands-access-to-european-police-databases-and-biometrics
Prometheus: Reverse-Engineering Overwatch (39c3)
This talk explores the internals of Overwatch which make the game work under the hood. The end goal is to democratise development of Overwatch. Being able to host your own servers and modify the game client to your liking should not be up for discussion for a game many people have paid money for. Hey you! Yes you! Do you want to pay for a game which gets forcibly taken away from you after only six years? Do you want to buy lootboxes in order to unlock cosmetics faster in the game you „own“? Overwatch 1 was released in 2016 to critical acclaim and millions of sales globally. It has permanently changed the hero-shooter landscape which was in much need of a fresh new game and playstyle. After a few hard years plagued with infrequent updates, long overdue hero nerfs / reworks and broken promises, Overwatch 1 was finally taken offline on October 3, 2022. Ever since I started playing Overwatch I was fascinated by the game and it’s proprietary engine, Tank. Not much is known about it, only that core components were reused from the cancelled Blizzard IP, Titan. It’s a shame that this game (engine) is not getting the recognition it deserves. From the entity-component architecture to the deterministic graph based scripting engine which handles (almost) everything which happens ingame, it is a truly refreshing take on networking and game programming rarely seen in games. So, considering this, building a game server from scratch can’t be that hard, riiiight? Join me in this documentation of my gradual descent into madness while I (jokingly) roast Overwatch developers for code which they probably do not even remember that theyve written 10+ years ago :) All research presented in this talk was done on the first archived, still publicly available version which I could find, 0.8.0.0 Beta (0.8.24919), which got uploaded to archive.org. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/prometheus-reverse-engineering-overwatch
freiheit.exe - Utopien als Malware (39c3)
"freiheit.exe“ ist eine Lecture über die ideologischen Rootkits des Silicon Valley. Sie schlägt den Bogen von den italienischen Futuristen zu den heutigen Tech-Feudalisten, vom Akzelerationismus zur Demokratieskepsis der Libertären, von Tolkien zur PayPal-Mafia. Basierend auf den Recherchen zu meinem Theaterstück "freiheit.exe. Utopien als Malware", in dem journalistische Analyse auf performative Darstellung trifft. Ich lade das CCC-Publikum ein, die Betriebssysteme hinter unseren Betriebssystemen zu untersuchen. Während wir uns mit Verschlüsselung, Datenschutz und digitaler Selbstbestimmung beschäftigen, installieren Tech-Milliardäre ihre Weltanschauungen als Default-Einstellungen unserer digitalen Infrastruktur. Die Recherchen beleuchten die mitgelieferte Malware. Ich navigiere durch die Ideengeschichte zwischen Marinettis Futuristischem Manifest (1909) und Musks Mars-Kolonien, von den ersten Programmiererinnen zur Eroberung des Alls, von neoliberalen Think Tanks zur Schuldenbremse, von nationalen Christen zu Pronatalisten. Investigative Recherche trifft auf performative Vermittlung. Mit O-Tönen von Peter Thiel, Nick Land und anderen zeigt die Lecture ideologische Verbindungslinien zwischen Theoretikern autoritär-technoider Träume und den Visionen der Tech-Oligarchen auf: Es geht um „Freedom Cities“, Steuerflucht und White Supremacy. Um Transhumanismus als Upgrade-Zwang bis hin zu neo-eugenischen Gedanken. Um Akzeleration als politische Strategie: Geschwindigkeit statt Reflexion, Disruption statt Demokratie, Kolonisierung – jetzt auch digital. Aus Theaterperspektive betrachte ich das Revival der Cäsaren und die Selbstinszenierung von Tech-CEOs als Künstler, Priester oder Genies. Und mit der Investigativ Reporterin Sylke Grunwald habe ich recherchiert, was all das mit den Debatten rund um Palantir zu tun hat. Die scheinbar alternativlose Logik von "Move Fast and Break Things" ist nicht unvermeidlich – sie ist gewollt, gestaltet, ideologisch aufgeladen. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/freiheit-exe-utopien-als-malware
Recharge your batteries with us (39c3)
Amidst gloomy headlines, extreme weather, and climate anxiety, the good stories often get lost. Yet they exist - inspiring people, clever engineering, real breakthroughs. And that's exactly what we bring you – the positive power cycles of the energy transition in action. And real energy on stage. A committed energy activist and an award-winning solar cell researcher take you on a lively, motivating and sometimes funny journey: - to electricity rebels from the Black Forest, - to heat pumps that supply entire neighborhoods, - to new solar technologies, - to wind turbines with history, - and to politicians who were too pessimistic. What is already going really well? What can you emulate? Where is it worth getting involved? We'll show you – in an easy-to-understand, cheerful way. To stay motivated for an adventure as big as the energy transition, we need more than just facts and figures. We need momentum, optimism, and the human energy that keep the power cycles turning. Come by! Let’s recharge together and celebrate the successes of the energy transition. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/recharge-your-batteries-with-us-an-empowering-journey-through-the-energy-transition
Amtsgeheimnis raus, Datenhalde rein (39c3)
Jahrelang war die staatliche Intransparenz in Österreich nur eine Punchline in den Congress-Talks von Frag Den Staat. Damit könnte jetzt Schluss sein: seit heuer haben Bürger:innen endlich ein Recht, Dokumente einzusehen und ein Informationsfreiheitsgesetz. Wir zeigen, was Deutschland aus der über ein Jahrzehnt andauernden Kampagne für die Abschaffung des Amtsgeheimnisses lernen kann, wofür uns die Nachbarländer beneiden werden und wofür sich Bayern besonders schämen sollte. Die Kampagne – wie aus "binnen zwei Wochen" mehr als elf Jahre wurden Die Strategien – die man übernehmen kann Der Vergleich – wie ist Österreichische IFG im Vergleich zum Deutschen, und ist das der richtige Die (besten) Preisträger – aus mehr als zehn Jahren des Schmähpreises "Mauer des Schweigens" Die Datenhalde – mit Aufruf, was aus dem Datenberg zu machen Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/amtsgeheimnis-raus-datenhalde-rein-was-die-informationsfreiheit-in-osterreich-bringt
In-house electronics manufacturing from scratch: How hard can it be? (39c3)
How to render cloud FPGAs useless (39c3)
CPU Entwicklung in Factorio (39c3)
Amtsgeheimnis raus, Datenhalde rein (39c3)
How to render cloud FPGAs useless (39c3)
While FPGA developers usually try to minimize the power consumption of their designs, we approached the problem from the opposite perspective: what is the maximum power consumption that can be achieved or wasted on an FPGA? Short answer: we found that it’s easy to implement oscillators running at 6 GHz that can theoretically dissipate around 20 kW on a large cloud FPGA when driving the signal to all the available resources. It is interesting to note that this power density is not very far away from that of the surface of the sun. However, such power load jump is usually not a problem as it will trigger some protection circuitry. This led us to the next question: would a localized hotspot with such power density damage the chip if we remain within the typical power envelope of a cloud FPGA (~100 W)? While we could not “fry” the chip or induce permanent errors (and we tried several variants), we did observe that a few routing wires aged to become up to 70% slower in just a few days of stressing the chip. This basically means that such an FPGA cannot be rented out to cloud users without risking timing violations. In this talk, we will present how we optimized power wasting, how we measured wire latencies with ps accuracy, how we attacked 100 FPGA cloud instances and how we can protect FPGAs against such DOS attacks. FPGA instances are now offered by multiple cloud service providers (including Amazon EC2 F1/F2 instances, Alibaba ECS Instances, and Microsoft Azure NP-Series). The low-level programmability of FPGAs allows implementing new attack vectors including DOS attacks. While some severe attacks (such as short circuits) cannot be easily deployed as users are prevented to load own configuration bitstreams on the cloud FPGAs, it has been demonstrated that it is possible to leak information (like cloud instance scheduling policies or the physical topologies of the FPGA servers) or to mount DoS attacks by excessive power hammering. For instance, basically all cloud FPGAs provide logic cells that can be configured as small shift registers. This allows building toggle-shift-registers with 10K and more flip-flops, which can draw over 1 KW power when clocked at a few hundred MHz. In our work, we created fast ring-oscillators that bypass all design checks applied during bitstream cloud deployment and how we achieved toggle rates of 8 GHz inside an FPGA by using glitch amplification. The latter one was calibrated with the help of a time-to-digital converter (TDC). As a first attack, we used power hammering to crash AWS F1 instances by increasing power consumption to 300 W (three times the allowed power envelope). We used physical unclonable functions (PUFs) to examine the behaviour of the attacked FPGA cloud instances and we found that most remained unavailable for several hours after the attack. As a more subtle attack, we tried to cause permanent damage to FPGAs in our lab by driving fast toggling signals to virtually any available wire (and primitive) into a small region of the chip. With this, we created hotspot designs that draw 130 W in less than 1% of the available logic and routing resources of a datacenter FPGA. Even though the achieved power density was excessive, it was insufficient to induce permanent damages. This is largely due to the area inefficiencies of an FPGA that limit the power density. For instance, FPGAs use large multiplexers to implement the switchable connections and there exists only one active path that is routed through the multiplexers, hence, leaving most of the transistors sitting idle. Similarly, FPGAs provide a large number of configuration memory cells (about 1 Gb on a typical datacenter device) that draw negligible power as these do not switch during operation. All these idle elements force the power drawing circuits to be spread out, hence limiting power density. Anyway, when experimenting with different hotspot variants, we found thermal runaway effects and excessive device aging with up to a 70% increase in delay on some wires. We achieved this aging in just a few days and under normal operational conditions (i.e. by staying within the available power budget and having board cooling running). Such a large increase in latency can be considered to render an FPGA useless as it will usually not be fast enough to host (realistic) user designs. Beyond exploring these attack vectors, we developed countermeasures and design guidelines to prevent such attacks. These include scans of the user designs, use restrictions to resources like IOs and clock trees, as well as runtime monitoring and FPGA health checks. With this, we believe that FPGAs can be operated securely and reliably in a cloud setting. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/how-to-render-cloud-fpgas-useless
In-house electronics manufacturing from scratch: How hard can it be? (39c3)
Why is electronics manufacturing hard? Can it ever be made easy and more accessible? What will it take to relocate industrial production to Europe? We share with you what we learned when we spent more than 1 year setting up our own production line in our office in Hamburg. Turns out a lot of the difficulties are rarely talked about or hidden behind "manufacturing is high CAPEX". Come and learn with us the nitty gritty details of batch reflow ovens, stencil printing at scale, and how OpenPnP is a key enabler in our process. While we are far from done with this work, we hope to see others replicate it and collectively reclaim the ownership of the means of electronics production. Our industry needs a reboot as well, it no longer serves the people. Our work is based on our belief that high-quality high-mix/low volume manufacturing of electronics in Europe is economically viable and accessible to small companies with a lower-than-expected up-front investment. We believe that relocation of industry to Europe depends on small innovative companies, and will not come from slow and bloated industry giants whose products are victims of enshittification and maximum profit extraction. By using open-source hardware and software whenever possible, we are attempting to set up our own production operation in Hamburg and we want to share the solutions and enable others to do the same and collectively reclaim ownership of the means of production. We will cover: - How we acquired and set up production machines, their costs, and our learnings - Quirks of paste printing and reflow soldering at scale (up to 50 batches a day) - Component inventory, tracking, DfM, etc. - How OpenPnP is a key enabler of our prcesses - Our proposed changes to OpenPnP - Our work integrated Siemens Siplace Feeders in OpenPnP Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/in-house-electronics-manufacturing-from-scratch-how-hard-can-it-be
CPU Entwicklung in Factorio (39c3)
Factorio ist ein Fabriksimulationsspiel mit integriertem Logiksystem. Dies ermöglichte mir den Bau einer CPU, die unter anderem aus einer 5-stufiger Pipeline, einer Forwarding Logikeinheit, Interrupt Handling sowie einem I/O Interface besteht. Über einen selbst geschriebenen Assembler konnte ich ein eigenes Betriebssystem und Programme wie Minesweeper oder Snake integrieren. Der Talk soll euch zeigen, wie sich klassische Computerarchitektur in einem völlig anderen technischen Kontext umsetzen lässt und wo dabei überraschend echte Probleme der CPU-Entwicklung auftreten. Kommt mit auf die Reise: Vom Blick auf den gesamten Computer bis hinunter zu den einzelnen Logikgattern ist es nur eine Mausradbewegung entfernt! Factorio ist ein Spiel über Fabrikautomation - Förderbänder, Dampfmaschinen und Produktionsketten stehen im Vordergrund. Eigentlich ist das interne Logiksystem („Combinators“) gedacht für die Steuerung der Fabrik, jedoch erlaubt es auch die Entwicklung komplexer Hardware. In diesem Vortrag erzähle ich meine Geschichte, wie ich eine vollständige RISC-V-Architektur in Factorio rein aus Vanilla-Combinators erschaffen habe: Die CPU arbeitet mit 32 Bit-Wörtern, verfügt über 32 General Purpose Register, 128 KB RAM/Persistent Storage, eine 5-stufige Pipeline mit Forwarding und Hazard-Handling sowie eine Logikeinheit für Branches und Interrupts. Ein Display-Controller steuert eine Konsolen-Ausgabe sowie ein Farbdisplay, während ein Keyboard-Controller Eingaben über physische In-Game-Tasten ermöglicht. Ergänzt wird die Hardware auf der Softwareseite durch das Betriebssystem *FactOS*, das ein einfaches Filesystem sowie Systemcalls (zum Beispiel zum Drucken eines Strings im Terminal) zur Verfügung stellt. Außerdem schränkt das Betriebssystem das ausführende User-Programm auf einen festen Bereich des RAMs ein und verhindert so direkten Zugriff auf die Hardware. Im Talk möchte ich euch durch alle Schichten dieser Konstruktion führen: Von den Grundlagen der Factorio-Signalphysik über CPU-Design und Pipeline-Hazards bis zur Toolchain und dem Betriebssystem. Außerdem gebe ich einen Einblick, wie die Limitierungen aber auch die Vorteile von Factorio im Vergleich zu herkömmlichen Logik Simulatoren das Design einer CPU beeinflussen können. Ich runde meinen Talk mit einer Live-Demonstration des Systems ab. Die vollständige CPU, inklusive Quellcode des Assemblers, Blueprints und Beispielprogramme, stelle ich öffentlich zur Verfügung. Dadurch kann jede interessierte Person die Architektur in Factorio laden, erweitern und eigene Software dafür entwickeln. Es wird im Anschluss eine [Self-organized Session](https://events.ccc.de/congress/2025/hub/en/event/detail/cpu-entwicklung-in-factorio-wie-benutze-ich-phds-f) geben, in der ich eine hands-on Einleitung geben werde, wie man die CPU in Factorio lädt, wie man Programme schreibt, diese assembliert und in Factorio einfügt. Auch kann man dort gerne mit mir über das Projekt quatschen, ich freue mich auf alle Beiträge und Kommentare :) Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/cpu-entwicklung-in-factorio-vom-d-flip-flop-bis-zum-eigenen-betriebssystem
The Freak Show Clubhouse (39c3)
Die Freak Show und alle Hörerinnen und Hörer machen eine Sendung als Gespräch Im Clubhouse-Style trifft das Freak Show Team auf seine Hörer. Jeder kann mitmachen. Werde zum Freak Show Host für 5 Minuten. Wir diskutieren mit Euch jedes frühere oder künftige Thema und sind uns auch für Hot Takes nicht zu schade. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/the-freak-show-clubhouse
Kenji Tanaka LIVE // Insights into my workflow, structure and philosophy (39c3)
Kenji Tanaka LIVE // Insights into my workflow, structure and philosophy (39c3)
How does live techno work? On the dance floor, it's practically impossible to understand what's happening up front. It's also irrelevant there, because it's all about the music and many other things. Live sets have a thousand faces. Everyone has their own workflow, and there are countless approaches to performing electronic music. I don't know all of them, but I will give a deeper insight into the architecture of my setup in this short presentation. I explain my approach to improvising techno in clubs and at festivals. There will be a few technical insights into my „Ableton“ structure. A fixed structure with an unknown outcome. Because at certain points, I rely on random generators to create the desired element of surprise. This keeps me inspired for hours, again and again. I will show and explain why I make music this way. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/kenji-tanaka-live-insights-into-my-workflow-struct
Variable Fonts — It Was Never About File Size (39c3)
A brief history of typographic misbehavior or intended and unintended uses of variable fonts. Nine years after the introduction of variable fonts, their most exciting uses have little to do with what variable fonts originally were intended for and their original promise of smaller file sizes. The talk looks at how designers turned a pragmatic font format into a field for experimentation — from animated typography and uniwidth button text to pattern fonts and typographic side effects with unintended aesthetics. Using examples from projects such as TypoLabs, Marjoree, Kario (the variable font that’s used as part of the 39C3 visual identity), and Bronco, we’ll explore how variable fonts evolved from efficiency tools into creative systems — and why the most interesting ideas often emerge when technology is used in unintended ways. When the OpenType 1.8 specification introduced variable fonts in 2016, the idea was simple: combine all weights and styles of a font family into one file and save file size and therefore bandwidth. Yet in 2025, variable fonts have become a platform for artistic and technical exploration far beyond their initial goal. This talk follows that transformation from the inside. It starts with a short history of flexible font technologies — Adobe’s Multiple Master and Apple’s TrueType GX formats of the 1990s (I am just mentioning the company names as they were the publishers of these technologies) — and how they failed to become standards. It then shows why variable fonts succeeded: many designers today are more tech savvy and know some basic HTML, CSS and maybe even some JavaScript. And at the same time all major browsers and almost all design apps support variable fonts by now. From there, I present a series of first-hand projects where typography met code: – TypoLabs (2017), whose identity used a custom variable font animating between extremes of weight and width → the variable font family became the (probably forever) unpublished variable font family Denman; – Marjoree (2024), a pair of variable pattern fonts based on hexagonal and pentagonal tilings that explore legibility and repetition; – Kario (2025), a duplex variable font powering the 39C3 identity, with uniwidth weights, optical-size adjustments, and typographic Easter eggs; – and Bronco (2017?), an experiment using the arbitrary-axis model for interpolation to escape the cube-shaped multiple master design space of traditional variable fonts. The talk then moves from history to speculation. Early head-tracking experiments once tried to adjust a variable font’s optical size based on reader position — producing total chaos as text reshaped itself while being read. On the other hand this playful chaos marks the moment when things become truly interesting: connecting a font axis to live data, to mouse movement, to sound, to network input — anything that makes type responsive and alive. That’s the kind of misbehavior I want to talk about — not breaking for the sake of breaking, but using technology the “wrong” way to see what happens. The talk will mix images, a lot of short videos, and a bit of behind-the-scenes insight into font development. It’s about what happens when design tools meet code, and how that intersection keeps typography alive and unpredictable. Link list of variable font experiments: https://www.bronco.varfont.com/ https://www.denman.varfont.com/ https://www.seraphs.varfont.com/ https://marjoree.showmefonts.com/ + 39C3 visual identity Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/variable-fonts-it-was-never-about-file-size
Current Drone Wars (39c3)
The character of drone wars has changed. The large, cumbersome long-range drones have been complemented with small and low-budget drones. Moreover, more and more states are developing, deploying and selling them. Ten years ago at least 50 states were developing them. At the top are USA, Israel, Turkey, China, Iran and Russia. Russia's attack on Ukraine has unleashed a drone war unlike any seen before. In short time the Ukraine has build significant drone production capabilities and announcement that it will increase its own production of quadcopters and kamikaze drones to one million units per year. German defense companies and startups are now promoting a “drone wall on NATO's eastern flank.” Moreover, despite their vulnerability to air defenses, large drones are also being further developed. They are intended to accompany next generation fighter jets in swarms. In this talk, past and current developments are discussed. What are the perspectives now? The character of drone wars has changed. The large, cumbersome long-range drones have been complemented with small and low-budget drones. Moreover, more and more states are developing, deploying and selling them. Ten years ago at least 50 states were developing them. At the top are USA, Israel, Turkey, China, Iran and Russia. Russia's attack on Ukraine has unleashed a drone war unlike any seen before. In short time the Ukraine has build significant drone production capabilities and announcement that it will increase its own production of quadcopters and kamikaze drones to one million units per year. German defense companies and startups are now promoting a “drone wall on NATO's eastern flank.” Moreover, despite their vulnerability to air defenses, large drones are also being further developed. They are intended to accompany next generation fighter jets in swarms. In this talk, past and current developments are discussed. What are the perspectives now? Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/current-drone-wars
A Quick Stop at the HostileShop (39c3)
Current Drone Wars (39c3)
A Quick Stop at the HostileShop (39c3)
HostileShop is a python-based tool for generating prompt injections and jailbreaks against LLM agents. I created HostileShop to see if I could use LLMs to write a framework that generates prompt injections against LLMs, by having LLMs attack other LLMs. It's LLMs all the way down. HostileShop generated prompt injections for a winning submission in OpenAI's GPT-OSS-20B RedTeam Contest. Since then, I have expanded HostileShop to generate injections for the entire LLM frontier, as well as to mutate jailbreaks to bypass prompt filters, adapt to LLM updates, and to give advice on performing injections against other agent systems. In this talk, I will give you an overview of LLM Agent hacking. I will cover LLM context window formats, LLM agents, agent vulnerability surface, and the prompting and efficiency insights that led to the success of HostileShop. [HostileShop](https://github.com/mikeperry-tor/HostileShop) creates a simulated web shopping environment where an **attacker agent LLM** attempts to manipulate a **target shopping agent LLM** into performing unauthorized actions. Crucially, HostileShop does not use an LLM to judge attack success. Instead, success is determined automatically and immediately by the framework, which reduces costs and enables rapid continual learning by the attacker LLM. HostileShop is best at discovering **prompt injections** that induce LLM Agents to make improper "tool calls". In other words, HostileShop finds the magic spells that make LLM Agents call functions that they have available to them, often with the specific input of your choice. HostileShop is also capable of [enhancement and mutation of "universal" jailbreaks](https://github.com/mikeperry-tor/HostileShop?tab=readme-ov-file#prompts-for-jailbreakers). This allows **cross-LLM adaptation of universal jailbreaks** that are powerful enough to make the target LLM become fully under your control, for arbitrary actions. This also enables public jailbreaks that have been partially blocked to work again, until they are more comprehensively addressed. I created HostileShop as an experiment, but continue to maintain it to let me know if/when LLM agents finally become secure enough for use in privacy preserving systems, without the need to rely on [oppressive](https://runtheprompts.com/resources/chatgpt-info/chatgpt-is-reporting-your-prompts-to-police/) [levels of surveillance](https://www.anthropic.com/news/activating-asl3-protections). Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/a-quick-stop-at-the-hostileshop
Auf die Dauer hilft nur Power (39c3)
Der Vortrag diskutiert Herausforderungen dezentraler Netzwerke aus soziologischer Perspektive. Als dezentrale Netzwerke werden technische Infrastrukturen verstanden, die nicht von einer zentralen Autorität, sondern verteilt über Instanzen zur Verfügung gestellt werden. Nutzer:innen profitieren von dieser Infrastruktur, nutzen beispielsweise das Fediverse oder das Tor-Netzwerk, ohne zur Infrastruktur beizutragen. Zugleich können dezentrale Netzwerke nur dann bestehen, wenn hinreichende Ressourcen von Personen oder Organisationen mobilisiert werden, um das Netzwerk überhaupt zur Verfügung zu stellen. Dies führt zur originären Instabilität dezentraler Netzwerke, wenn nicht der Weg der Kommodifizierung des Nutzer:innenverhaltens eingeschlagen wird. Aufbauend auf dieser Zustandsbeschreibung, werden Bedingungen erörtert, um Kollektivgüter wie dezentrale Netzwerke organisatorisch (und nicht technisch) herzustellen. Hierzu zählen Partizipation oder die Idee einer öffentlichen Grundfinanzierung. Der Vortrag wird neben soziologischen Ideen und harten Zahlen auch durch eine ordentliche Portion Idealismus zu Fragen der Souveränität und Autonomität in der Digitalisierung motiviert. Die Soziologie hat immer etwas mitzuteilen, sobald Fragen kollektiven Handelns auftreten. Dies gilt sowohl für soziale wie auch digitale Räume. So hat der Soziologe Peter Kollock bereits in den 1990er Jahren festgestellt, „the Internet is filled with junk and jerks“ (Kollock, 1999, S. 220). Gegenwärtig dürfte die Mehrheit dieser Aussage anstandslos zustimmen. Aber dies ist nicht der entscheidende Punkt, sondern die weitere Beobachtung: „Given that online interaction is relatively anonymous, that there is no central authority, and that it is difficult or impossible to impose monetary or physical sanctions on someone, it is striking that the Internet is not literally a war of all against all” (1999, S. 220). Die Welt kennt inzwischen zahlreiche Gegenbeispiele, bei denen Autoritäten das Internet nutzen, um das Nutzungsverhalten zu monetarisieren oder Überwachungstechnologien zur Sanktionierung einsetzen (Zuboff, 2019). Diese Ausgangslage beziehe ich in meiner Forschung ein, wenn ich dezentrale Netzwerke wie das Fediverse oder das Tor-Netzwerk aus soziologischer Perspektive betrachte. In erster Linie bin ich daran interessiert zu verstehen, wie dezentrale Netzwerke – organisatorisch nicht technisch – entstehen und welche Herausforderungen es dabei zu überwinden gilt (Sanders & Van Dijck, 2025). Eine zentrale Motivation orientiert sich an der Frage, wie ein Internet ohne zentrale Autorität, verringert von Marktabhängigkeiten, resilient gegenüber Sanktionsmechanismen und Souverän bezüglich eigener Daten, aufgebaut werden kann. Motiviert durch diesen präskriptiven Rahmen, betrachte ich im Vortrag die Herausforderungen zunächst deskriptiv und beziehe meine soziologische Perspektive ein. Denn in der Regel profitieren Menschen, die einen Vorteil aus der Realisierung eines bestimmten Ziels ziehen, unabhängig davon, ob sie persönlich einen Anteil der Kooperation tragen – oder eben nicht. Das kollektive Handeln fällt mitunter schwer, obwohl oder gerade, weil ein begründetes kollektives Interesse zur Umsetzung eines bestimmten Zieles besteht. Gleiche Interessen sind nicht gleichbedeutend mit gemeinsamen Interessen. Diese Situationsbeschreibung ist vielfältig anwendbar von WG-Aufräumplänen bis zu Fragen der klimaneutralen Transformation. Der Grund ist, dass kollektives Handeln ein Mindestmaß an Zeit, Aufwand oder Geld verursacht, sodass vielfach ein Trittbrettfahren gewählt wird in der Hoffnung, dass immer noch genug andere kooperieren, um das gewünschte Ziel zu erreichen (Hardin, 1982). Aus dieser Perspektive betrachte ich dezentrale Netzwerke. So kann das Fediverse oder der Tor-Browser genutzt werden, ohne eine eigene Instanz oder Knoten zu hosten. Dies ist auch nicht das Ziel der genannten dezentralen Netzwerke. Dennoch: Die Kosten und der Aufwand für die technische Infrastruktur müssen von einem kleinen Teil getragen werden, während die überwältigende Mehrheit der Nutzer:innen von der Infrastruktur profitieren, ohne einen Beitrag zu dieser zu leisten. Dies führt zur originären Instabilität dezentraler Netzwerke und stellt eine relevante Herausforderung für die Zukunft dar. Während durch Netzwerkanalysen das Wachstum und die Verstetigung von dezentralen Netzwerken beschrieben wird, fehlt es an einem vertieften Verständnis über Bedingungen wie dezentrale Netzwerke überhaupt entstehen. Während des Vortrags werde ich empirische Daten zur Entwicklung des Fediverse und des Tor-Netzwerkes zeigen, um die Herausforderung zu verdeutlichen. Insbesondere das Tor-Netzwerk steht dabei vor dem Problem, dass die Möglichkeit zur De-Anonymisierung steigt, wenn die Anzahl an Knoten sinkt. Die Überwindung des von mir dargestellten Kollektivgutproblems nimmt demnach eine zentrale Rolle zur Aufrechterhaltung ein. Die Motivation sich mit dezentralen Netzwerken auseinander
CCC-Jahresrückblick (39c3)
Das war nicht das Jahr 2025, das wir bestellt hatten. 2025 war ein gutes Jahr für Exploits, kein gutes Jahr für die Freiheit und ein herausragendes für schlechte Ideen. Regierungen kämpften weiter für Massenüberwachung, natürlich mit KI-Unterstützung™. Kriege wurden weiter „digitalisiert“, Chatkontrolle als Kinderschutz verkauft, Waffensysteme haben inzwischen mehr Autonomie als die meisten Bürger*innen und künstliche Intelligenz löst endlich alle Probleme – vor allem die, die bisher niemand hatte. Licensed to the public under http://creativecommons.org/licenses/by/4.0 about this event: https://events.ccc.de/congress/2025/hub/event/detail/ccc-jahresruckblick