Security

Apple Bug Bounty: AI Reports and the 180-Day Pause

#Apple#Security#AI Adoption

Apple security bounty program guidelines on AI-found reports

Apple Security Bounty is the program that pays researchers for reporting vulnerabilities in Apple products and services. Its official guidelines mention AI in several places.

They cover which reports earn a reward, what happens when a researcher keeps sending reports that do not, and how to write a report. The guidelines say the following about AI:

  • Issues discovered by AI without proper validation count as infeasible reports, which are ineligible for a reward.
  • Submitting ineligible reports repeatedly can lead to a 180-day pause on processing.
  • Researchers with more than two paused periods may be permanently removed.
  • A report should avoid lengthy descriptions generated by AI tools.

Apple bug bounty rewards and payouts for unvalidated AI findings

A report on the ineligible list earns no bounty reward. Among the ineligible kinds are infeasible ones: theoretical issues, or issues discovered by AI without proper validation.

The list has a second entry: reports that are incomplete or not actionable are also ineligible, even if you were the first to report. The guidelines give one example: a report without a reliable way to reproduce the issue.

Infeasible reports, such as reports of theoretical issues or issues discovered by AI without proper validation. ／ Reports that are incomplete or not actionable, even if you were the first to report — such as reports without a reliable way to reproduce the issue. — From the list of reports ineligible for a reward
View official source →

Apple security bounty program pause: 180 days after repeated reports

Apple may pause processing for 180 days if you repeatedly submit ineligible reports. The guidelines again name two kinds of infeasible report as examples: reports about theoretical issues, and issues discovered by AI without proper validation.

Two words in the text set the limits of the rule. The condition is "repeatedly", so one ineligible report is not described as a trigger. The verb is "may", so a pause is a possible response, not an automatic one.

Pauses also add up. Researchers with more than two paused status periods may be permanently removed from the Apple Security Bounty program.

Apple states its reason in the same part of the guidelines. It receives many reports that claim serious security or privacy issues but were generated by LLMs (large language models, the AI systems behind chatbots).

These reports arrive without the proof or validation by a person that the program requires. Investigating these and other ineligible reports can keep Apple from acting quickly to resolve serious and critical security issues.

If you repeatedly submit ineligible reports — including infeasible reports about theoretical issues or those discovered by AI without proper validation — we may pause processing your reports for 180 days. Researchers with more than two paused status periods may be permanently removed from the Apple Security Bounty program. ／ We receive many reports claiming to be about serious security or privacy issues, but that are generated by LLMs and submitted without the required proof or validation by a person. Investigating these and other ineligible reports can prevent us from acting quickly to resolve serious and critical security issues. — From the pause condition and the stated background
View official source →

Apple Target Flag exception for paused security bounty researchers

While your status is paused, you are ineligible for bounty rewards and for credit in security advisories. Apple does not process your reports during that time. There is one exception, for a report that clearly demonstrates the security or privacy issue in one of two ways:

  • By capturing the applicable Target Flag.
  • With a fully packaged virtualization of iOS or macOS.
If your status in Apple Security Bounty is paused, you are ineligible for bounty rewards or credit in security advisories during this time. We will not process your reports, unless your report clearly demonstrates the security or privacy issue by capturing the applicable Target Flag or with a fully packaged virtualization of iOS or macOS. — From what applies during a pause
View official source →

Apple bug bounty program report criteria and who earns rewards

A report needs a working exploit or a reliable proof of concept (PoC). A PoC is a demonstration that the issue is real. The exploit has to explain the conditions required to start the attack, and the control, data or privilege you gain by the end of it.

At a minimum, the report includes a concise, numbered list of the steps required to reproduce the issue. The guidelines add an instruction about style: avoid lengthy descriptions generated by AI tools.

The reward goes to one report per issue. Only the first complete and actionable one Apple receives for an issue is eligible, even if it is not the first or only one received.

A working exploit that explains the conditions required to start the attack and the control, data, or privilege you gain by the end of the attack — or a reliable proof of concept (PoC) for the issue you’re reporting. Your report must at least include a concise, numbered list of steps required to reproduce the issue. ／ Avoid lengthy descriptions generated by AI tools. ／ Only the first complete and actionable report we receive for an issue is eligible for a reward, even if it’s not the first or only report we receive. — From the report requirements, the note on AI-generated descriptions, and reward eligibility
View official source →

Apple submission cap reports versus the security bounty guidelines

9to5Mac reported on August 3, 2026 that Apple confirmed a cap and a 30-day cool-off period on submissions through its security portal. The article (9to5Mac) credits the Financial Times report as its source.

The guideline passages quoted above contain neither a submission cap nor a 30-day period. The 180-day pause in those passages is a separate measure. It applies to repeated ineligible reports, not to the number of reports a researcher sends.

In a statement to the FT quoted by 9to5Mac, Apple said it had adjusted the number of new reports a researcher can have open at once. Researchers can request an increase to that limit at any time, according to the company.

For a separate case where an AI-assisted result passed an automated check despite being wrong, see Collatz Conjecture Proof: Why Lean Accepted a Fake. The case is unrelated to Apple's program but shows the same need for a person to validate AI output.

Apple has confirmed to The Financial Times that it has “introduced a cap and a 30-day cool-off period on submissions through its internal security portal, requiring users to submit requests for an increased quota.” ／ we recently adjusted the number of new reports a researcher can have open at once. ／ The company added that researchers “can easily request an increase to that limit at any time to ensure critical reports reach our security teams.” — From 9to5Mac's report on the cap and Apple's statement
View official source →

FAQ

Does the Apple bug bounty accept reports of issues found with AI?
Apple's guidelines list issues discovered by AI without proper validation as infeasible reports. Infeasible reports are ineligible for a reward. The same list names theoretical issues.
When does Apple pause processing your security bounty reports?
Apple says it may pause processing for 180 days if you repeatedly submit ineligible reports. That includes infeasible reports about theoretical issues and issues discovered by AI without proper validation. Researchers with more than two paused status periods may be permanently removed from the program.
What happens to your reports during a pause?
You are ineligible for bounty rewards and for credit in security advisories, and Apple does not process your reports. The exception is a report that clearly demonstrates the issue by capturing the applicable Target Flag or with a fully packaged virtualization of iOS or macOS.
What must an Apple security bounty report include?
A working exploit or a reliable proof of concept. The exploit explains the conditions required to start the attack and what you gain by its end. At minimum, the report includes a concise, numbered list of steps to reproduce the issue, and Apple asks you to avoid lengthy descriptions generated by AI tools.