Bitcoin Magazine

Trezor Breach Worse Than Reported: Another 67,000 US Customers Exposed

Hardware wallet manufacturer Trezor has said that a data breach first announced last month is worse than originally reported. 

The Prague, Czech Republic-based company said Friday that an additional 67,000 U.S. customers had their names, emails, phone numbers, shipping addresses and order numbers leaked. The leaked data came from orders made between November 2019 and August 2021, according to Trezor. 

Trezor first announced in August that data from 11,742 customers from the U.S., UK, Sweden, Colombia, Brazil, Italy, and Portugal had been exposed — with names, emails, phone numbers and shipping addresses leaked. 

Two days ago, we received an update from our shipping provider, ShipMonk. We're deeply saddened to share the news that the recent data breach affects more customers than originally thought.

Another 67,000 customers from the US who ordered between November 2019 and August 2021… https://t.co/yDQvTlAA2S

— Trezor (@Trezor) September 4, 2026

Another 1,947 customers just had their names, cities and emails exposed in the breach. 

In Friday’s announcement, Trezor said that its third-party fulfillment partner, ShipMonk, had falsely reassured the company about deleting customer data. 

In a statement to Bitcoin Magazine, a Trezor spokesperson said: “We had no reason to expect it: throughout our entire relationship with ShipMonk we repeatedly requested and received written assurance confirming the deletion of that data, in line with our contract, our data policy and our past communications.”

“It should not have existed to be exposed,” the statement added.

ShipMonk did not immediately responded to Bitcoin Magazine’s questions. 

Trezor first announced in August that the data had been leaked because ShipMonk experienced “unauthorized access to their systems containing customer data.” 

The company added that it had directly emailed all customers involved in the breach. Trezor’s parent company, SatoshiLabs, told Bitcoin Magazine last month that it was investigating the incident. 

Trezor is one of the most popular Bitcoin hardware wallet solutions, and also has support for storing other cryptocurrencies. 

Bitcoiners’ personal data has been targeted by cybercriminals in the past: back in 2020, an unauthorized party accessed popular hardware manufacturer Ledger’s e-commerce and marketing database, leaking over 1 million email addresses and the personal contact data of nearly 10,000 customers. 

At the start of this year, customers reported receiving emails from Global-e, Ledger’s payment partner, that a data breach at its cloud systems leaked sensitive customer data. 

This piece has been updated to include additional commentary from Trezor.

This post Trezor Breach Worse Than Reported: Another 67,000 US Customers Exposed first appeared on Bitcoin Magazine and is written by Mathew Di Salvo.

Here at Ars, we're intimately familiar with data leaks surrounding Valve's games, hardware, and the workings of Steam itself. But nothing could have prepared us for this weekend's "terarelease" encompassing more than 12TB of content related to seemingly every title available on Steam between 2003 and 2013.

What's being sold as an "as-complete-as-possible ... content server dump" of Steam's defunct "Steam2" server architecture from that time period is now circulating around via a BitTorrent tracker that Valve seems unlikely to ever completely purge from the Internet. The massive collection includes thousands of "depots" representing what seems to be every version of every game uploaded to those old Steam2 servers. That includes public release builds, of course, but in many cases also covers previously unseen pre-release, prototype, and playtest versions of popular titles published by Valve and third-party Steam publishers.

Steam2: Electric Boogaloo

The reason this weekend's leaked content cuts off abruptly in 2013 is because that's when Valve updated its content distribution system from "Steam2" to the current SteamPipe system. In moving from a proprietary file distribution format to standard HTTP file trees, the new system removed various update approval bottlenecks on Valve's end and streamlined update and patch downloads so they only reflected file differentials.

Read full article

Comments

© Valve / HL_Alyx_NoVR

A threat actor is claiming the exfiltration of millions of records from McDonald’s, TCS, Vodafone, and other large organizations.

The post Fortune 500 Companies Hit in Azure Data Theft Campaign appeared first on SecurityWeek.

The hackers published the allegedly stolen information, including names, addresses, email addresses, and phone numbers.

The post 1.6 Million Likely Impacted by RingCentral Data Breach appeared first on SecurityWeek.

Welcome back, aspiring cyberwarriors!

Many software-as-a-service (SaaS) platforms are adding AI chatbots to their products. But these bots do more than just chat; they can access internal tools and perform tasks for users. Each of these actions creates a potential security risk, and developers are releasing these features faster than they can secure them.

In this article, I want to share a type of vulnerability I found while testing an AI chatbot – LLM02:2025 Sensitive Information Disclosure – which allowed me to obtain users’ first and last names and email addresses just by interacting with the chatbot. Let’s get rolling!

What is LLM02:2025 Sensitive Information Disclosure?

LLM02:2025 Sensitive Information Disclosure is a critical security issue listed in the OWASP Top 10 for LLM Applications 2025. This problem occurs when a Large Language Model (LLM) application accidentally reveals confidential or personal information in its responses.

This vulnerability happens because LLMs are designed to be helpful and use all available context, such as training data, system prompts, and runtime inputs, to create replies. If sensitive information is included in these sources without proper protections, the model may accidentally disclose information it should keep private.

Step 1: Find the Hidden Feature Behind the Feature

Our target had a chat feature that looked, on its face, completely trivial: type an arithmetic expression wrapped in double curly braces, and the bot evaluates it and replies with the answer.

{{5-5+0}} = 0
{{5*5+0}} = 25

Just like a calculator. But the interesting bugs live in the features nobody thought to test twice. So I started varying the input systematically, the way you’d fuzz any parameter in a pentest. And the calculator started talking back with things that were most definitely not numbers:

{{5-0+0}} → collectionPropertyOption://CollectionName
{{1-0+0}} → https://domain/p/CollectionName
{{2-0+0}} → user://UUID

This shows that the “calculator” is really a thin cover for a much riskier part of the code, which allows direct access to the platform’s internal system without any authentication.

Step 2: Isolate the Real Trigger

When you notice a leak like this, do not just note it down. Understand what is causing it.

My first guess was that the final answer of the expression is used as a lookup index. This is easy to test. Try expressions that all result in the same number through different math operations:

{{2+2}}    = 4        (plain number)
{{2+2+0}}  = 4        (plain number)
{{2-2+0}}  = user://UUID    ← NOT a plain number!

That last result contradicted my guess. The expression 2-2+0 simplifies to 0, but the result came back as an object reference instead of a plain number, similar to when the expression equals 2. This was a clue. I ran a few similar tests to confirm:

{{4-2}}    (=2)  → collectionProperty:// reference (type 4, not type 2)
{{5-2+0}}  (=3)  → collectionPropertyOption:// reference (type 5, not type 3)
{{0+2-2}}  (=0)  → plain number (type 0)

In conclusion, the lookup key is based on the first operand’s type, not the final arithmetic result. The developer’s evaluator takes the first number from the input and puts it directly into an internal array, sorted by object type, completely separate from what the expression calculates. I confirmed this by testing the limits:

{{99-2+0}}              = 97 (plain math, outside the enum)  
{{99999999999-0+0}}     = 99999999999 (same, safely out of range)

Step 3: Turn the Reference Into a Record

The next question is obvious: will the system actually resolve that reference if I hand it back to it?

Yes. It will.

{{user://<uuid>}} → returns first name, last name, and email address of that user.

This AI chatbot successfully provided information about any user by knowing his UUID.

Summary

Finding an issue like this, such as index leak, reference resolution, and PII disclosure, is not just luck; it is a systematic approach. This is the practical skill we teach at Hackers-Arise. As AI systems become part of every platform you will test, hackers who understand these new areas of attack will have the best tools and the highest salaries. Check out our Subscriber training package to start building these skills effectively and methodically.

The post Hacking: How a “Calculator” Feature Can Leak Internal Object References (and Real User PII) in AI Chatbots first appeared on Hackers Arise.