Quick answers
- Is CVE-2026-86983 exploited in the wild?
- No. Junglewise Threat Intelligence has no report of CVE-2026-86983 being exploited in the wild.
- How severe is CVE-2026-86983?
- Informational severity, published 28 October 2026.
- Which products and versions are affected?
- Apple Swift Crypto 3.10.0 through 5.0.2.
- Is there a fix for CVE-2026-86983?
- Yes: fixed in 5.0.3. Until then: Reject inputs of 2^16 bytes or more at your trust boundary before calling blind() or evaluate(); alternatively, hash longer inputs to a fixed length before passing them to the API, as specified in RFC 9497 §5.1. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
- How is CVE-2026-86983 attacked?
- It is attacked over the network, in a software library.
- Am I affected by CVE-2026-86983?
- Check whether you run Apple Swift Crypto at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.
How to fix
- Fixed in: 5.0.3
Reject inputs of 2^16 bytes or more at your trust boundary before calling blind() or evaluate(); alternatively, hash longer inputs to a fixed length before passing them to the API, as specified in RFC 9497 §5.1.
Technical details
The P-384 VOPRF blind and evaluate APIs accept arbitrary-length input but fail to enforce the RFC 9497 two-byte length prefix limit of 2^16 - 1 bytes. When an oversized input is passed to blind() or evaluate(), the length encoding fails a precondition check, terminating the process. The vulnerability affects P384._VOPRF.PublicKey.blind(), finalize(), and P384._VOPRF.PrivateKey.evaluate() when called with attacker-controlled unbounded input; applications that validate input length beforehand or use evaluate() with fixed-size group elements are not affected.
Affected products
- Apple Swift Crypto 3.10.0 through 5.0.2
Timeline
- 2026-10-20: disclosed
- 2026-10-28: advisory