Noise from invalid or low-impact reports creates a burden for programs and reduces the time that can be spent on reports that matter. With Human-Augmented Signal, analysts from HackerOne review reports flagged with a high noise probability by utilizing various criteria to classify all incoming reports automatically. This best-effort human review reduces the risk of false positives. Any report dismissed by HackerOne analysts will appear in your inbox as Not Applicable but will not generate a notification. While eliminating all noise is improbable, Human-Augmented Signal aids in helping to provide a 90%+ signal experience for all programs.
- HackerOne's machine learning technology identifies reports that have a high chance of being invalid.
- Reports that are flagged are reviewed by a HackerOne Security Analyst. The analyst looks at these factors to determine whether a report requires review:
- Does the hacker meet the signal requirements of the program?
- Did the hacker select a weakness?
- Does the report contain abusive language?
- Does the report match earlier reported issues?
- If a report is determined to be invalid, it'll be closed as Not Applicable.
Note: If a hacker comments on their report marked as N/A, the report will reopen so that additional comments can be addressed.
HackerOne's systems are constantly improving to detect and eliminate noise across the platform. Don't hesitate to reach out if you have feedback on the common noise you see in your program.
No. By enabling this feature, you grant authorized HackerOne staff members temporary access to only flagged reports for the sole purpose of providing this service. Our Human-Augmented Signal services are governed by the same Terms & Conditions you have with HackerOne.
To enable Human-Augmented Signal:
- Go to your program's Program Settings > Program > Hacker Management > Submission.
- Select Enabled.