‼️ BREAKING: Our investigations team at Duel is in contact with the Revolut hacker, and we've found out a lot more about how he did what he did. - The hacker got access to government employee accounts using an infostealer. After gaining access to an employee's email, they would log in, add their own recovery email, start to log everything, and silently listen in. - To not get caught, they would instantly delete any email sent that was not intended for the original employee. The inbox was checked 24/7 for any new response to the hacker's sent emails. Upon receiving one, the hacker would instantly download it as a .eml and delete it before the original employee noticed - The hacker says that at first he used to forge court orders (presumably when targeting other companies), but quickly realised that this wouldn't work for Revolut. After some research, they decided the best entity to target was Revolut's Lithuania-based Revolut Bank UAB, which would respond to a European Investigation Order (as shown in the images). - With this stolen email, the hacker sent one request, originally 5 months ago, attached below. Revolut believed it, thinking it was the Italian government, and complied with the order. - From the hacked email, the hacker was able to control and end emails that looked like they genuinely came from multiple Italian government email addresses. - The hacker continuously sent out requests over the course of 5 months. Not once did Revolut ask questions or not send over the information. In one incident, the hacker accidentally sent the wrong document. Instead of realising what was going on, Revolut's support guided them on what to change (shown in an image below) We remain in contact with the hacker and we've requested exclusivity of information related to the story to be kept with Duel. We believe it is in the public's best interest for EVERY piece of information related to this to be released, so that the extent of Revolut's failure can be brought to light, as well as the sheer stupidity of the manner in which the KYC paradigm is currently conducted. The Duel team hopes that Revolut will be held accountable for their lack of due diligence and betraying their customers in such a severe manner, especially given the breadth and depth of the breach. Lives are now at risk. I'm personal friends with one of the victims, and he'll probably have to move houses due to the continued (credible) kidnap threats. We hope to soon release a much more detailed article with more information, emails, screenshots and more.

Meet Piecrust, a Dusk community builder behind five applications and seven AI model repositories. @ruskvm’s work spans a DEX, two wallets, a block explorer and a .dusk web resolver. Here’s what he built 👇

We have designated ChatGPT as a Very Large Online Search Engine and Reddit and Roblox as Very Large Online Platforms under the Digital Services Act. They now have four months to comply with additional DSA obligations. More: link.europa.eu/8bmk3M #DSA

We’ve written an FAQ to answer some of the questions we've received about watermarking. In summary: • We’re implementing watermarking to comply with the EU AI Act. Other major model developers have signed the same Code of Practice and will also be implementing watermarking; • Our watermarking method doesn’t have any practical impact on the quality or content of Claude’s outputs; • The difference between watermarked and un-watermarked text will not be distinguishable to readers; • Nothing is added to the text and there are no hidden characters; • Watermarking doesn’t require extra tokens, and will not be more expensive; • Watermarks can’t be traced to a specific person, organization, or chat. Read more: anthropic.com/news/claude-te…

🚨 JUST IN: Claude models will now have invisible watermarks embedded in ALL text, and ALL metadata attached to files…

Everything consumer AI still collapses down to companionship. Everyone hates hearing this. It reeks of male-coded dystopian loneliness. But AI can only deliver on its consumer promises if it can be a good friend. Trusted, warm, intimate. Personality is the holy grail.

📢Meet Qwen3.8-Max — our most capable model to date. Next week, the open weights of Qwen3.8-Max will be released, and Qwen3.8-27B is also going open-weights to meet you all!🎉 Qwen3.8-Max, a new bar for coding and cowork at 2.4T parameters: - Autonomous coding: 10+ days of self-evolving development, from empty folder to production without hand-holding, complete project trace in the GitHub：github.com/qwen-code-dev-bot… - Real work, real results: Production-quality deliverables across hundreds of professions. - Long-horizon mastery: System-level autonomous planning with closed-loop adaptive learning, driving 500+ turns of chip design optimization and 365 days of e-commerce strategy. - Native multimodal intelligence: Vision isn't just input — it's a continuous feedback loop for planning, execution, and self-correction. 💰Pricing: Input: $2.0 / M tokens Output: $6.0 / M tokens Implicit Caching: $0.25 / M tokens Start building with Qwen3.8-Max! 🚀 📖 Blog: qwen.ai/blog?id=qwen3.8 ✅ Qwen Studio: chat.qwen.ai/?models=qwen3.8… ⚡ API: qwencloud.com/models/qwen3.8…

Currently I'm using on 3 projects: - Cursor + Grok 4.5 - Kimi CLI with Kimi K3 - Codex with GPT 5.6 Sol - Hermes with GLM 5.2, MiniMax M3 and others - Claude with Fable for some complex reviews OpenCode and omp on other projects too. It seems impossible, but I'm starting to experience a new kind of flow, where I talk with each agent, swap between them, let them interact with each other through me... And I feel lost in this status for hours. And at the end? I get an incredible amount of work done! Including very complex one. WOW! Initially I was feeling exhausted by this, but now, week after week I start to feel this more natural, it's becoming the new norm. Isn't this happening to you too anon?

1/ Announcing OpenDusk A community-led initiative, accelerating growth in the Dusk ecosystem.