
Railway breach leads to $740,000 in fraudulent AI charges
Attackers stole environment secrets from at most about 100,000 workspaces after two infrastructure flaws.
78 stories, newest first · one card per story

Attackers stole environment secrets from at most about 100,000 workspaces after two infrastructure flaws.

Version 3.9 tests 59 agents on 327 challenges with a 1,000-second budget and deterministic grading.

Wiz plans to start an external ISMAP assessment in early 2027 after a completed gap analysis.

Green Agent identifies issue owners, root causes, and generates environment-specific fixes to reduce mean time to remediation.

Customers must assume environment variables were decrypted; Northflank has not said how attackers got in or how many were affected.

Of 100 security leaders, 55% run autonomous agents and 52% found unauthorized AI apps in the past year.

Schwartz says AI tooling found bugs that sat in PHP for more than a decade, and some sites are still unpatched.

Cloudflare saw millions of attempts within hours, while Calif, OpenAI and Wiz say AI helped turn long-known PHP issues into practical exploits.

The post explains finding enrichment, deduplication and attack-path scoring inside Wiz Code, not a new incident.
Wiz says unauthenticated attackers can run code or read process memory that may hold passwords and tokens.

Calif and Wiz found the bugs; OpenAI helped PHP develop patches, and the NSA urged taking unpatched systems offline.

OpenAI says the models found the issues before attackers, and a GitHub tool checks whether a PHP version is affected.

OpenAI and Wiz say the holes were found before attackers; unpatched sites should go offline, and PHP 8.5.12 is recommended.

Calif used GPT-5.6 Sol and Wiz independently reported both flaws; maintainers led the fixes and a version checker is available.

GKE Standard and Autopilot can enable the sensor with one flag; credentials stay in Secret Manager.

Wiz says notable supply-chain attacks rose from about 10% of significant incidents in late 2025 to 1 in 4.

One confirmed entry was Java remote code execution in an exposed Kafka UI, followed by malware and command-and-control.

A 2026 buyer's comparison ranks 11 IaC security scanners, platforms, and Terraform governance tools.

Wiz promotes ASPM connectors that centralize scanner findings and prioritize exploitable application risks.

Wiz makes its AI security agents available in FedRAMP High Wiz for Gov for U.S. agencies.

A 2026 ranking puts Wiz, Snyk IaC, and Checkov atop ten infrastructure-as-code security tools.

A 2026 buyer’s guide ranks twelve ASPM platforms, led by Apiiro, ArmorCode, and Cycode.

Wiz says Red Agent found a public sensitive-data page at an unnamed Fortune 500 financial firm.

Wiz launched AI SAST in public preview to find code flaws traditional scanners miss.
Today 1.5K readers · 2.4K views · 754 reads · all time 189K views