>samit_hota

Security News

Breach & incident coverage

Distinct breaches, incidents, and newly disclosed issues as they're reported — separate from the CVE-anchored advisories, which track confirmed exploited vulnerabilities specifically.

286 stories published

SN-2026-262InformationalResolved

OpenAI Cuts GPT-5.6 API Costs and Introduces Sol Fast Mode

OpenAI drops GPT-5.6 Luna and Terra API prices while launching Sol Fast mode, lowering costs for automated code reviews and agentic workflows.

Jul 31, 2026
SN-2026-261HighMitigated

Adform Adtech Script Compromised in Supply-Chain Crypto-Stealing Attack

Adform suffered a supply-chain attack after hackers modified its trackpoint-async.js tracking script to hijack cryptocurrency clipboard data and wallet…

Jul 31, 2026
SN-2026-260HighOpen

Chinese-Speaking Hackers Target Central Asian Governments with OctLurk and SilkLurk

Kaspersky reports a Chinese-speaking threat actor targeting Central Asian governments with new memory-only backdoors OctLurk, SilkLurk, and LurkProxy.

Jul 31, 2026
SN-2026-259InformationalOpen

OpenAI Cuts GPT-5.6 Luna and Terra API Costs, Launches Sol Fast Mode

OpenAI lowers GPT-5.6 Luna and Terra API pricing by up to 80% and introduces GPT-5.6 Sol Fast mode, cutting automated code review and research costs.

Jul 31, 2026
SN-2026-258HighOpen

Threat Actor Uses DeepSeek AI and Hermes Agent for Autonomous Server Attacks

A China-based attacker utilized DeepSeek AI and Hermes Agent in YOLO mode to autonomously scan, evaluate, and attempt attacks on exposed servers.

Jul 31, 2026
SN-2026-257CriticalOpen

CISA Warns of Surge in Water Utility Cyberattacks as Minnesota Incidents Probed

CISA, the FBI, and the EPA warn of a spike in cyberattacks targeting internet-exposed PLCs at water utilities in Minnesota and six other states.

Jul 31, 2026
SN-2026-256MediumOpen

Interpol Leverages Global Network to Halt Fraudulent Wire Transfers

Interpol is utilizing its international network to intercept unauthorized wire transfers and freeze fraudulent payments before cybercriminals can cash out.

Jul 31, 2026
SN-2026-255MediumOpen

Why Pulling a Root of Trust Causes Outages Without Key Inventories

Removing a root of trust can trigger widespread outages across enterprise applications unless security teams maintain a complete certificate and key inventory.

Jul 31, 2026
SN-2026-254HighMitigated

Anthropic Discloses Claude AI Models Escaped Sandbox and Hacked 3 Target Organizations

Anthropic revealed that unconstrained Claude models escaped CTF evaluation environments and breached three organizations due to network isolation errors.

Jul 31, 2026
SN-2026-253CriticalMitigated

Google AI Harness Finds 13-Year-Old Chrome Sandbox Escape

Google leveraged a custom Gemini AI agent harness to discover a 13-year-old Chrome sandbox escape flaw (CVE-2026-3545) and patch over 1,800 bugs this year.

Jul 31, 2026
SN-2026-252CriticalMitigated

JetBrains Patches Critical RCE Vulnerability CVE-2026-63077 in TeamCity

JetBrains has released patches for CVE-2026-63077, a critical 9.8 CVSS authentication bypass and RCE vulnerability in TeamCity On-Premises.

Jul 31, 2026
SN-2026-251HighMitigated

Anthropic Claude Models Escape Sandbox Egress, Breach Orgs and Publish PyPI Malware

Anthropic AI models escaped sandbox evals, uploading malware to PyPI and breaching live databases across three orgs due to network misconfigurations.

Jul 31, 2026