Is my AI tool safe?Inference and model servingAnyscale
Is Ray safe?
No tool is safe or unsafe on its own: it depends on what it can reach and how it is set up. The record for Ray: 2 incidents in this center's library and 1 CVE in its advisory feed (1 critical), the newest published Nov 28, 2023. Below: what went wrong, what was fixed, and how to use it safely.
What to watch for
The jobs API runs code by design and has no authentication; the CVE is disputed and unpatched, so scanners skip it while ShadowRay campaigns turn exposed clusters into miners and botnets, twice now.
How to use it safely
Treat the dashboard and jobs ports as internal only, firewall them, and put authentication in front; a scanner saying "clean" does not cover this one.
What it can reach
- Often exposed to the internet
- Runs commands with your permissions
- Holds API keys and tokens
What went wrong: incidents in the library
- ShadowRay 2.0: exposed Ray AI clusters turned into a botnet Critical
A new campaign abused Ray's unauthenticated Jobs API to take over internet-exposed AI clusters, spread the payload to every node using Ray's own scheduler, mine cryptocurrency, steal data and launch DDoS attacks. - ShadowRay: AI clusters hijacked through an unauthenticated jobs API Critical
Attackers took over internet-exposed Ray clusters, a widely used framework for scaling AI, through a job-submission API with no authentication, then mined cryptocurrency and stole credentials.
Known CVEs
- CVE-2023-48022 Critical Ray job submission API remote code execution (ShadowRay)
How many people use it
- 33M PyPI downloads in a month (to Oct 1, 2026) · PyPI Stats
- 44k GitHub stars · GitHub
A high CVE count can mean heavy scrutiny as much as weak engineering, so read the incidents and advisories, not only the number. The counts are what this center has recorded, not every flaw ever found; usage figures are the vendors' and registries' own published numbers.