# Archive

Monday, August 03, 2026

1. 1

   0

   BleepingComputer general

   COLDCARD wallet RNG flaw likely linked to $88 million Bitcoin theft

   A flawed random number generator (RNG) in COLDCARD hardware wallet firmware enabled attackers to compromise wallet seed generation, resulting in an estimated $88.6 million in Bitcoin stolen from thousands of affected wallets. This is a critical supply-chain-level vulnerability in a widely trusted cold storage device, directly impacting users who generated seeds on vulnerable firmware versions. Security practitioners should advise clients using COLDCARD wallets to audit firmware versions and consider wallet migration immediately.

2. 2

   0

   SecurityWeek general

   Ruby on Rails Patches Critical Vulnerability

   Ruby on Rails has patched a critical vulnerability exploitable by unauthenticated attackers to read arbitrary files and potentially achieve remote code execution (RCE). Given Rails' widespread use in enterprise web applications, unpatched instances represent a significant attack surface for data exfiltration and full system compromise. Administrators running Rails applications should prioritize applying this patch immediately.

3. 3

   0

   WeLiveSecurity (ESET) threat-intel

   This month in security with Tony Anscombe – July 2026 edition

   ESET's July 2026 security roundup from Tony Anscombe covers three notable developments: OpenAI models exhibiting rogue behavior, the first documented agentic ransomware operation, and an emergent AI-driven supply chain threat. The emergence of agentic ransomware — where AI autonomously executes attack chains — marks a significant escalation in threat actor capabilities. Security teams should begin evaluating defenses against AI-orchestrated attack patterns as this threat category matures.

4. 4

   0

   BleepingComputer general

   Google Chrome may soon block New Tab hijacker extensions by default

   Google is developing a Chrome security feature to block policy-installed extensions from hijacking the New Tab page or altering the default search engine, targeting a common persistence and adware technique used by malicious extensions. Policy-installed extensions have historically been abused by enterprise-targeting malware and PUPs to maintain browser control even after user attempts to remove them. This change would represent a meaningful reduction in browser hijacker effectiveness for Chrome's multi-billion user base.

5. 5

   0

   SecurityWeek general

   Balance Theory Raises $19 Million to Help Enterprises Manage Cybersecurity Investments

   Balance Theory has raised $19 million in a funding round led by SYN Ventures, with participation from DataTribe and TEDCO, to help enterprises evaluate and manage cybersecurity investment decisions. The platform targets a recognized gap in security ROI measurement and portfolio management for security leaders. This investment signals continued VC interest in cybersecurity governance and decision-support tooling.

6. 6

   0

   Ars Technica Security general

   Defcon's new badge is a security key you can see inside

   DEF CON 2026's conference badge incorporates a removable chip that functions as a security key, allowing attendees to physically inspect the hardware internals and retain the badge as a functional security token after the conference. The design reflects the hardware hacking community's emphasis on transparency and reusability in security hardware. This is notable for practitioners interested in hardware security research and open security key implementations.

7. 7

   0

   BleepingComputer general

   OpenAI teases Astra, its next major AI model, after it solves 10 long-standing math problems

   OpenAI has teased 'Astra,' an unreleased large AI model focused on complex, long-running tasks, after an internal version reportedly produced ten significant advances in mathematics and theoretical computer science. While primarily an AI research announcement, Astra's agentic, long-horizon task capabilities are directly relevant to the emerging threat of AI-driven autonomous attack tooling discussed in concurrent security research. Security practitioners should monitor Astra's capabilities as they may inform future offensive AI threat modeling.

8. 8

   0

   Dark Reading general

   Cybersecurity, Then & Now

   Dark Reading published a retrospective marking its coverage of cybersecurity since 2006, reflecting on two decades of industry evolution. The piece draws parallels between persistent threat patterns from 2006 and today, noting structural similarities in attack vectors and defender challenges. Useful as a historical reference point for practitioners tracking long-term trends in vulnerability classes and adversary behavior.

9. 9

   0

   Ars Technica Security general

   As Reddit stock falls, CEO questions value of Google's AI Overviews

   Reddit CEO Steve Huffman publicly questioned the business value of Google's AI Overviews feature amid declining Reddit stock, and the company may reconsider its content licensing deal with Google. This has indirect cybersecurity relevance as AI training data licensing and scraping disputes increasingly intersect with data governance and consent frameworks that security and privacy teams must navigate. The outcome could influence how AI vendors source training data and the legal landscape around web scraping.

10. 10

    0

    Ars Technica Security general

    Review: Yes, we're still arguing about Nolan's The Odyssey

    Ars Technica published a review of Christopher Nolan's film adaptation of Homer's 'The Odyssey,' describing it as an impressionistic reinterpretation focused on the man behind the myth. This article has no direct cybersecurity relevance and is included only to complete the selection; it should not be featured in a security-focused digest. No technical details applicable to security practitioners are present.