Monday 16 November
Global

Your region, kept on this device.

Orca Research finds widespread critical flaws in AI-generated apps and packages

Severity
High
Sources
1

Orca Research Pod reports that 51% of organisations use AI to build software and 81% have critical vulnerabilities in AI packages. CVE-2025-48757 involved a missing database policy in AI-generated projects, which exposed data across more than 170 production apps. The post argues that these shadow AI-generated apps often sit outside governance. Defenders should locate them in cloud accounts and in code already scanned, and check database access policies on Supabase-backed apps.

For security analysts · Written by AI from 1 source · Wrong?

What to do

  1. Search cloud accounts and code repositories for AI-generated apps that teams may have deployed without review.
  2. Check that Supabase projects have database access policies configured, since a missing policy caused the CVE-2025-48757 exposure.

Written by AI from the sources, checked against them

Full story timeline · 1 update

  1. Tue 10 NovemberShadow AI Generated Apps: Find Them in the Cloud and Code You Already ScanOrca Security Blog

Who reported · 1 outlet

Orca Security Blog

Who and what is in it

Companies

Report a problem