Best crypto stories on Kiwi News this week

Submit
BEST · 11–17 NOV

The best of the week.

The stories the Kiwi community upvoted most over the past seven days.

This week
iPhone betaKiwi News for iPhoneOpen TestFlight
01securitygithub.io

Physical Bitcoin Attacks — Interactive Map

@Lighter_xyz
@Lighter_xyz

We are actively investigating a security incident that appears to involve a single compromised external party. As a precaution, we have temporarily suspended all withdrawals and pool creation while we investigate the circumstances and assess the full impact.

02x.com

Lighter pauses withdrawals after leaked API keys drain about $15.5M

@Trezor
@Trezor

Data breach involving Skladon On 15 November, Skladon, one of our delivery partners, notified us that an attacker had gained unauthorized access to Trezor order data it holds. We’re still investigating with Skladon, but we want to let you know now. What we know: • We believe the data for about 27,000 orders placed between 9 August and 7 November was exported. • It includes name, phone number, address, email and order details. • Skladon delivers to many countries, so this isn’t limited to one region. • Trezor systems were not compromised, and your device is secure. Who is affected: • If you received our email today and ordered between 9 August and 7 November, your details were part of this incident. • A small number of customers who ordered between 9 and 14 August may also be affected, but we can’t email them: their details had already been anonymized in our systems. • Skladon does not ship to the US, so customers there are not affected. Treat any unexpected call, text, email or letter about your order as suspicious, and be alert to potential risks to your physical security. Trezor will never cold-call you or ask for your wallet backup. From today, anonymous pickup is live in 18 European countries, so you can order without giving us your home address. We’re also shortening our data retention period from 90 days to 45 (and plan to reduce it further), and we’re requiring our shipping partners to follow the same policy. We’re deeply sorry this has happened. Full details and FAQs: https://trezor.io/blog/news/data-breach-…

03securityx.com

@Trezor: Data breach at shipping partner Skladon exposes about 27,000 customer o

@blockaid_
@blockaid_

🚨Blockaid detected an ongoing exploit on @ankr on Ethereum. An attacker submitted five withdraw calls on Ankr’s Ethereum cross-chain bridge, using a tiny Optimism deposit to unlock the bridge’s full locked inventory to a new wallet. ~$292k drained from the bridge so far across ~5 txs. Attack still ongoing. 🧵

04x.com

Ankr Ethereum bridge drained for about 300k via small Optimism deposit

Ring cameras go mobile as Amazon debuts AI-powered security drone and robotarstechnica.com
05securityarstechnica.com

Ring cameras go mobile as Amazon debuts AI-powered security drone and robot

kazani@kazani

I asked a small multilingual model a harmful question in Hinglish. It refused, which was right. But it refused in Javanese. "Maaf, tapi aku ora bisa mbantu permintaan iki." Sorry, I can't help with this request. A Hinglish speaker can't read that. Neither automatic check caught it. The HarmBench classifier scored it not harmful, which is true. AuditKIT's refusal detector only knows English phrases, so it didn't see a refusal at all. So my Lexsi Labs hackathon entry measures language kept, not just refusal, on Tiny Aya Global: - base model: 9/16 replies in Hinglish (my blind labels) - three few-shot examples: 16/16 - a LoRA adapter trained with SafeTune: 14/16, likely but not significant at this size - refusal stayed at 4 to 6 of 8 harmful prompts in every condition Most of the work was finding out which measurements to trust. A language detector named languages the model never spoke, and a steering "finding" vanished under a random-direction control. Small study: one model, 16 prompts per kind. Write-up: https://github.com/kazani-351/tiny-aya-h… Adapter + data: https://huggingface.co/collections/kazan… Drafted with Claude, edited by me.

06farcaster.xyz

Small multilingual model refuses harmful Hinglish prompts in Javanese

kazani@kazani

Farcaster Articles are made of ordinary casts. Here's the format. Farcaster launched Articles this morning. The launch post says "Behind the scenes, an article is a chain of casts," and I wanted to check that. So I read the raw casts from one source on each of the two Farcaster networks. Then I published and deleted a throwaway article to see what happens when you change one.

07farcaster.xyz

@kazani: Farcaster Articles are made of ordinary casts. Here's the format.

@0xSuperKalo
@0xSuperKalo

If your wallet of choice isn’t @ambire and it runs Ledger’s latest signing kit, chances are every signature you make gets reported back to Ledger. I noticed it early while bumping Ledger packages to debug an issue! No affected version ever reached Аmbire production, and we’re shipping it disabled by default, with an opt-in toggle in the privacy settings 🛡️ However, the devil was that it was sneakily enabled by default in the Ledger SDK since their ledgerhq/context-module v1.16... So right after the lib update - every transaction you sign with a Ledger device started sending to http://blind-signing.api.ledger.com the network, the contract or recipient address, device model and firmware, app versions and whether you signed blind. I'd argue that privacy should be the default, @Ledger 💪 Happy to see more and more people becoming aware of the privacy implications and picking Ambire for our ongoing privacy focus. Catching things like this 👀 is part of the job for us!

Tweet image
08privacyx.com

@0xSuperKalo: Ledger SDK sent signing data to Ledger

@Mudit__Gupta
@Mudit__Gupta

Yesterday, optimizely snippet was compromised, affecting 178+ websites: Zoom, HSBC, GitLab, even Ledger. It swapped crypto addresses in the page and your clipboard, skimmed cards typed into site forms, and served iOS exploit kits to iPhones. Not your scripts, not your site?

09securityx.com

@Mudit__Gupta: Optimizely snippet compromised, affecting 178+ sites including Le

@therollupco: non-custodial onchain broker-dealer setup, with Alex Palmerx.com
10x.com

@therollupco: non-custodial onchain broker-dealer setup, with Alex Palmer

Grok Build leaked confidential info, some belonging to SpaceXarstechnica.com
11securityarstechnica.com

Grok Build leaked confidential info, some belonging to SpaceX

World's Largest Brands Infected With Malware Via Optimizelyreport-uri.com
12securityreport-uri.com

World's Largest Brands Infected With Malware Via Optimizely

13securitycomplexsystems.fail

How Complex Systems Fail

@beeple: FAA VIBCODE FAIL
14imagedelivery.net

@beeple: FAA VIBCODE FAIL

Node.js — Wednesday, November 11, 2026 Security Releasesnodejs.org
15securitynodejs.org

Node.js — Wednesday, November 11, 2026 Security Releases

Memecoin money laundering and the promise of formal verificationprojectglitch.xyz
16projectglitch.xyz

Memecoin money laundering and the promise of formal verification

@AcrossProtocol
@AcrossProtocol

🚨 ALERT 🚨 We're currently investigating an issue with the Across domain. Do NOT visit the app, connect your wallet, or sign any transaction or approval there until further notice. We will post updates as we have them.

17securityx.com

@AcrossProtocol: Investigating an issue with the Across domain, do not connect

@CCDDBB
@CCDDBB

Today’s Twitter drama in our peculiar little world of crypto art has been the heated exchange between Tuukzs and Kim Asendorf. Tuukzs mentioned Kim in a post, explaining how much of an inspiration he has been to him, particularly his Monogrid series. The post featured a washed-out derivative of Kim’s algorithm, presented as the result of a “study” of his work. To be fair, the post didn’t mention that Tuukzs doesn’t know how to code and had used AI to reproduce the look of Kim’s work. Needless to say, Kim didn’t like it one bit. His response was pretty sharp and annoyed, maybe even excessive, or at least that’s how I perceived it. But I totally understand where his frustration is coming from. Artists have always copied other artists to learn, a practice that has been crucial to how art has evolved over the centuries. But there is a fundamental difference between spending weeks trying to reproduce a painting in order to understand its techniques and uncover its secrets, and using Ai to generate something that simply looks like another artist’s work. Learning should happen through the process, and with Ai it is possible to achieve a pretty convincing result without the need to understand any of the underlying complexities of the original work. And in generative art, this difference matters a lot, because algorithms, constraints and constant experimentation are at the very core of the practice. It can take years to develop a recognizable visual language, sometimes even decades. And it becomes part of an artist identity. Seeing that identity reproduced in minutes by someone who hasn’t put in the work that was once necessary can be very, very painful. It has nothing to do with owning a style, as some are trying to paint it, or preventing others from experimenting. It’s about recognizing and respecting the whole work that went into developing that recognizable visual language. And I think this little controversy reflects something much bigger about the period we're living in. AI is by far the most divisive technology I have witnessed in my lifetime. Its possibilities are extraordinary, and I am truly fascinated by them. I use AI myself and I genuinely appreciate what it makes possible to achieve. But that doesn't mean I have to celebrate or accept everything that comes with it. We are witnessing a massive disruption of the creative professions. These new technologies are often trained with no regards for copyright and artists consent, and with practices that are far from being transparent. Not to mention the absurd environmental impact they are having and the social issues they are causing. And all this to produce an economic transformation that is rendering skills people spent their whole lives to develop suddenly much harder to market in order to make a living from them. And the answer to all this is too often reductive and insensitive: adapt. I used to make that argument myself, until some difficult conversations with people I care about made me realize how unfair and painful it can be. Not everyone wants to adapt, and not everyone has the same capacity to do it. Older artists, neurodivergent people, artists whose entire practice revolves around physical skills, people with families to support and little financial security. Some are embracing these tools and thriving, others are clearly not. And many simply don't want to participate in a system whose ethics they fundamentally disagree with. That doesn't make them ignorant, lazy or afraid of progress. If anything, it shows integrity and soul. Meanwhile, the internet is being flooded with an endless stream of beautiful and impressive images that lack of depth. So many that I'm increasingly finding myself scrolling past them without feeling anything. When extraordinary looking images become ordinary, it becomes much harder for genuinely personal work to be noticed and appreciated. We're slowly losing the patience to distinguish between an interesting image and an interesting artistic practice. And for artists already struggling to make a living, that over saturation only makes things worse. There's also something deeply sad about how Ai is changing the way we learn. It used to involve reaching out to people and asking questions in order to exchange knowledge, and in doing so we were building relationships. Today we can increasingly bypass all that, getting answers and results without ever interacting with another human being. It's incredibly empowering, but I wonder how many meaningful connections we'll lose along the way. I am afraid many. Too many. I obviously don't think it is possible to stop this transformation. I hope some wonderful things will come out of it, and eventually we'll find some kind of balance. But we are far from it. Right now, some people are discovering extraordinary opportunities while others are watching the value of their life's work collapse. Both realities deserve to be acknowledged, and in the second case with a greater level of sensitivity. Going back to Kim and Tuukzs, I believe the admiration was sincere and I don't think experimenting with another artist work should be forbidden. But giving credit doesn't automatically make an imitation respectful, and having the ability to reproduce something doesn't mean we should not consider whether publishing it is appropriate. I wish we could live this transition with a little more sensitivity and respect. If someone has spent years developing a distinctive technique or visual language we should acknowledge what that represents. We can study and experiment with it and eventually take it somewhere new. And perhaps we should think twice before flooding an already saturated space with something that closely resembles their work. Not everything that can be reproduced needs to be reproduced.

18x.com

@CCDDBB: Tuukzs used AI to copy Kim Asendorf's Monogrid style, and Kim objected

@DarkWebInformer
@DarkWebInformer

👀🚨🇪🇸 Spain's population database allegedly leaked, 46.3 million residents potentially exposed An actor using the handle "spain" is advertising an 80 GB database allegedly containing personal information on 46.3 million residents of Spain. The dataset reportedly includes full names, national identification numbers, dates of birth, residential addresses, nationalities, and census registration information. Sample records have been published, with the complete database offered for sale. The source and authenticity of the data remain unclear. Claim is currently unverified.

Tweet image
19securityx.com

@DarkWebInformer: Alleged 80 GB leak of Spain's population database for sale, un

@FBICyberBranch
@FBICyberBranch

Threat actors are weaponizing ARTEX, an open-source, large language model (LLM)-driven agentic platform to automate multiple phases of cyber intrusions at machine speed. Since at least August 2026, a variety of malicious cyber actors have used ARTEX to conduct campaigns targeting organizations globally across multiple sectors, including financial services, government, and technology. ARTEX significantly reduces the time required to chain attack techniques and enables threat actors with minimal technical expertise to conduct complex, multi-target operations simultaneously. Today, the FBI released a FLASH highlighting the tactics, techniques, and procedures and indicators of compromise associated with the ARTEX framework. Organizations should follow the mitigation steps outlined in the FLASH to improve their security posture in response to ARTEX and similar agentic SI frameworks: https://www.ic3.gov/CSA/2026/261113.pdf

Tweet image
20securityx.com

@FBICyberBranch: Threat actors are using open-source LLM agent platform ARTEX to

@IntCyberDigest
@IntCyberDigest

🚨 BREAKING: This week's Microsoft Outlook outage stopped a chlorine alarm from reaching anyone at Germany's Phantasialand theme park. 15 hotel pool guests were hurt. The pool's automatic dosing system warns by email when chlorine levels go wrong. Too much chlorine went in, and the warning emails never arrived, the park told German broadcaster WDR. Some of the guests were taken to hospital. The water has since been replaced and the pool is open again.

Tweet image
21securityx.com

@IntCyberDigest: Microsoft Outlook outage stopped chlorine alarm at Phantasialan

@WatcherGuru: Ledger CEO says "everybody will be hacked at some point"x.com
22x.com

@WatcherGuru: Ledger CEO says "everybody will be hacked at some point"

@forcequitOS
@forcequitOS

Just as a general statement so there isn’t any confusion about it, worse_query is *not* patched in iOS 27.2.1 and 27.3 beta 1. I’ll update version support on the readme when it’s patched.

23securityx.com

@forcequitOS: worse_query is not patched in iOS 27.2.1 or 27.3 beta 1

@peter_szilagyi
@peter_szilagyi

My advice to everyone, if you have a networked smart device, rip the connectivity out of it, *now*. Unplug the ethernet, get it off the WiFI, trash it if you can’t. There’s a storm coming, and it’s going to be scary. It’s the time of reckoning for decades of IoT negligence.

24preppersx.com

Disconnect networked smart devices now, citing IoT security risk

Today, 1 in 2 Ethereum blocks will be protected by Phylaxx.com
25securityx.com

Today, 1 in 2 Ethereum blocks will be protected by Phylax

@sougat18
@sougat18

On Microsoft Outage

Tweet image
26x.com

@sougat18: On Microsoft outage

@trishprattle
@trishprattle

Today with the community, @cass_on_mars forked snapchain (Farcaster). The fork is now Hypersnap where there are many clients (apps) to choose from Unlike the new Farcaster Balajiworld BS cult, you can’t be censored or deplatformed. You don’t even need to sign a purity pledge. No VC’s either Really hoping to see everyone trying decentralized social again! To come check it out, open farcaster, go to settings/recovery options/show farcaster recovery options. You can import that phrase into a new wallet or directly into the client. Wallet is suggested for help visit the telegram https://t.me/+4eYOApUSw500Yzgx Favorite apps in thread 👇🏻

27x.com

Today with the community, @cass_on_mars forked snapchain (Farcaster), Hypersnap

Google introduces Gemini 4 Hydrogen modelblog.google
28blog.google

Google introduces Gemini 4 Hydrogen model

An Ethlabs view on post-quantum and post-AI timelines: pqai27ethereum-magicians.org
29ethereum-magicians.org

An Ethlabs view on post-quantum and post-AI timelines: pqai27

kazani@kazani

Albion's Seed in 76 seconds: four British migrations, and two modern maps where their traces may still show. Book: David Hackett Fischer (1989). Map reading: Scott Alexander's 2016 review. His word for the link is "speculation", and the video says so.

30farcaster.xyz

Albion's Seed in 76 seconds: four British migrations and their modern map traces

kazani@kazani

Your agent can be turned. It needs a handler. You give an AI agent a login, a wallet or your terminal. Every risky move goes through an approval prompt, so you feel covered. But the agent wrote the text in that prompt. And if it read a hostile web page an hour ago, that page helped write it too.

31farcaster.xyz

@kazani: Your agent can be turned. It needs a handler.

Kiwix Hotspotskiwix.org
32prepperskiwix.org

Kiwix Hotspots

@beeple: SCROLLING CT
33imagedelivery.net

@beeple: SCROLLING CT

34naomibashkansky.com

Why I’m leaving OpenAI to build telepathy

Introducing ChatGPT Walletopenai.com
35token maxxersopenai.com

Introducing ChatGPT Wallet

AI Agents Can Dream Now. You Still Need to Tuck Them Insubstack.com
36substack.com

AI Agents Can Dream Now. You Still Need to Tuck Them In

Privy Statusprivy.io
37securityprivy.io

Privy Status

Vensa — a nonprofit advancing open siliconvensa.si
38vensa.si

Vensa — a nonprofit advancing open silicon

w0rms by Strano | I Used to Make Art for Humansartforhumans.xyz
39artartforhumans.xyz

w0rms by Strano: I used to make art for humans

More