Weekly Cybersecurity Breach Roundup: DOGE Data Exposure, CIRO Phishing Attack, and Rising Threats This week’s cybersecurity landscape saw multiple high-profile incidents, including unauthorized data sharing by the U.S. Department of Government Efficiency (DOGE), a massive phishing breach in Canada, and a surge in critical vulnerabilities. ### U.S. DOGE Staff Exposed Social Security Data via Unauthorized Cloudflare Server Federal prosecutors confirmed that staff from Elon Musk’s Department of Government Efficiency (DOGE) uploaded sensitive Social Security Administration (SSA) data to an unauthorized Cloudflare server in March 2025. The breach, first reported by a whistleblower in August, involved employees sharing data via third-party links between March 7 and 17. The SSA remains uncertain whether the data was removed from Cloudflare. The incident is part of ongoing litigation over DOGE’s activities at the SSA, which critics claim wasted $21.7 billion. Prosecutors also revealed that a DOGE employee signed an agreement with a political advocacy group seeking voter fraud evidence, potentially linking SSA data to voter rolls. Two DOGE employees were referred to the U.S. Office of Special Counsel for possible Hatch Act violations, which prohibit federal employees from partisan activities. Additionally, a DOGE team member sent an encrypted file believed to contain names and addresses of 1,000 individuals to the Department of Homeland Security and a DOGE advisor at the Department of Labor. The SSA has been unable to decrypt the file. Another DOGE employee continued accessing the "Numident" database containing Social Security card applications and death records despite a court order revoking access. ### Canadian Investment Regulatory Organization (CIRO) Phishing Breach Affects 750,000 Investors The Canadian Investment Regulatory Organization (CIRO) disclosed a phishing attack in August 2025 that exposed sensitive data of approximately 750,000 investors. Compromised information includes names, contact details, dates of birth, Social Insurance numbers, government-issued IDs, investment account numbers, and account statements. CIRO confirmed that login credentials, passwords, and security questions were not accessed. ### UK NCSC Warns of Rising Russia-Aligned Hacktivist DDoS Attacks The UK’s National Cyber Security Centre (NCSC) issued an alert about increased denial-of-service (DDoS) attacks by Russian-aligned hacktivist groups, including NoName057(16). Targets include government bodies, local authorities, and critical infrastructure operators. The NCSC advised organizations to strengthen defenses with traffic filtering, web application firewalls, and rate-limiting policies. ### Ingram Micro Ransomware Attack Exposes 42,000 Employee Records IT distributor Ingram Micro suffered a July 2025 ransomware attack by the SafePay gang, which stole 3.5 terabytes of data, including names, birthdates, Social Security numbers, passport details, and employment records. The breach affected 42,521 individuals. Ingram took systems offline to contain the attack, causing service disruptions before restoring operations by July 9. SafePay later published the stolen data after Ingram refused to pay the ransom. ### CVE Disclosures Surge 21% in 2025 Vulnerability disclosures reached 48,185 in 2025 a 20.6% increase from the previous year with 3,984 critical and 15,003 high-severity flaws. December alone accounted for 5,500 CVEs, while February 26 saw a record 793 disclosures in a single day. Nearly 30% of exploited vulnerabilities were weaponized within one day of disclosure, and 25.8% lacked analysis in the National Vulnerability Database, complicating mitigation efforts. ### SK Telecom Challenges $91 Million Data Leak Fine South Korea’s SK Telecom is contesting a $91 million fine the largest ever imposed by the country’s privacy watchdog after a 2025 data breach exposed all 23 million of its mobile subscribers. The delayed disclosure led to a broader investigation, prompting SK Telecom to offer free USIM replacements. A ransomware group, CoinbaseCartel, later claimed responsibility, alleging it stole source code, project files, and AWS keys via a compromised Bitbucket account. ### Critical Chainlit Vulnerabilities Expose AI Data and Cloud Infrastructure Security researchers at Zafran Labs disclosed two critical flaws in the open-source AI framework Chainlit (CVE-2026-22218 and CVE-2026-22219). The vulnerabilities allow arbitrary file reads and server-side request forgery (SSRF), enabling attackers to access sensitive data, including AI prompts and credentials, and probe internal networks. Chainlit released patches to address the issues. ### North Korean Hackers Abuse Microsoft VS Code for Malware Delivery North Korean threat actors expanded their "Contagious Interview" campaign, using Microsoft Visual Studio Code to execute malware via malicious Git repositories. Victims are tricked into opening projects that automatically run attacker-controlled commands, deploying the EtherRAT macOS trojan. The group has also leveraged developer-friendly platforms like Vercel for command-and-control infrastructure.