Top Bug Bounty Programs 2026: Compare Rewards & Scope
This document lists 13 active bug bounty programs from major tech companies including Apple, Google, Meta, Microsoft, Coinbase, OpenAI, and others, offering rewards for security vulnerability discoveries. Researchers should apply by identifying a program matching their skills, reviewing the official scope and eligibility rules on the linked program pages, and submitting reports through each program's designated channels. Deliverables include valid security reports with reproduction steps, impact analysis, and proof of concept, following each program's specific guidelines. Applications are made directly via the official program URLs provided in the document.
Eligibility: Open to security researchers who comply with each program's rules, including scope, safe harbor, and reporting requirements; some programs have age (18+) and geographic restrictions.
Pipeline
Funder track record · does Multiple (Apple, Google, Meta, Microsoft, Coinbase, OpenAI, Intel, Anthropic, Shopify, GitHub) actually pay grants?
⚪ No payout history found · checked 2026-11-05 · 0 forum reports, 3 web pages read, 7 unverifiable claims dropped
Proven payer: at least 3 named grantees and one payout in the last 90 days. Every grantee below is quoted word for word from its source. When a source does not date a grant, the date is when that page or post was published.
Trust breakdown (2/3 checkable facts verified)
| Signal | Weight | Result | Note |
|---|---|---|---|
| deadline | 0.35 | 0% | rolling claimed but not stated |
| amount | 0.25 | 100% | figure appears in source |
| funder | 0.075 | 100% | named in source |
| source quality | 0.15 | 50% | social or search hit |
| model confidence | 0.1 | 90% | model self-report |
Fit breakdown (against the ORG_* profile)
| Signal | Weight | Result | Note |
|---|---|---|---|
| model fit | 0.5 | 10% | model rated 10% |
| ecosystem overlap | 0.25 | 0% | none vs aleo |
| category overlap | 0.15 | 17% | security vs zk, privacy, infra, dev_tooling, security, interoperability |
| tech overlap | 0.1 | 0% | none vs rust, leo, zk |
Sources (1)
| Role | Document | From | Fetched |
|---|---|---|---|
| primary | Top Bug Bounty Programs 2026: Compare Rewards & Scope | HackWednesday | Web discovery: ecosystem sweep | 2026-10-01 |
History (1)
| v | Changed | Fields |
|---|---|---|
| 1 | 2026-10-01 | created |
Extracted text
Bug Bounties # Top bug bounty programs 2026: compare rewards and scope. Find a program that fits your skills, not just the biggest reward. Compare research areas and advertised ceilings, then verify the official scope, eligibility and reporting rules before you test. Quick answerApple, Google, Meta, Microsoft, OpenAI, CoinbaseVerification dates listed per program [Compare 13 programs](#current-bug-bounty-rewards)[Check eligibility first](#bounty-eligibility)  Headline rewards are useful signals, but actual payouts depend on exploitability, scope, severity, and report quality. Choose your research path ## Scope first. Reward second. Start with a product you understand. Check that your target and testing method are permitted, read exclusions and safe-harbor terms, and plan how to demonstrate impact without accessing other people's data. The list is a research shortlist, not a payout ranking. Match your skillsClear scope mattersRead rules first Current rewards table ## Compare bug bounty rewards and AI research scope. Use this table as a starting point, then click through to the official program page before testing. Reward ceilings are not guaranteed payouts. All programsAI programs only Showing 13 programs in the table. Check individual eligibility before testing. Program Advertised reward ceiling Best fit Official source **Apple Security Bounty** Mobile / platform Rewards reaching $2,000,000; potential maximum above $5,000,000 with bonuses Advanced exploit chains, Apple platform research, Lockdown Mode, PCC, mobile security [Apple Security Bounty categories](https://security.apple.com/bounty/categories/) Last verified: September 5, 2026 **Google Android and Devices VRP** Mobile / Android Up to $1,500,000 for zero-click full-chain Pixel Titan M2 compromise with persistence Pixel, Android, firmware, hardware-backed security, persistence, zero-click research [Google: Android and Chrome VRPs for the AI era](https://bughunters.google.com/blog/evolving-the-android-chrome-vrps-for-the-ai-era) Last verified: September 5, 2026 **Coinbase Bug Bounty** Crypto / fintech Extreme-severity reports remain eligible for up to $1,000,000 Crypto, fintech, account protection, high-impact Web2 and Web3 security research [Coinbase: bug bounty focus in the age of AI](https://www.coinbase.com/blog/focusing-our-bug-bounty-program-on-what-matters-most-in-the-age-of-ai) Last verified: September 5, 2026 **Google Chrome VRP** Browser Up to $250,000 for full-chain browser-process exploits; additional MiraclePtr bypass rewards can apply Browser exploitation, sandbox escapes, memory safety, MiraclePtr bypass research [Google Chrome VRP rules](https://bughunters.google.com/about/rules/chrome-friends/chrome-vulnerability-reward-program-rules) Last verified: September 5, 2026 **Meta Bug Bounty** Web / platform Up to $300,000 for listed top mobile RCE category, before possible bonuses Large-scale web, mobile, account security, platform abuse, and privacy-impact research [Meta Bug Bounty](https://www.facebook.com/whitehat/) Last verified: September 5, 2026 **Microsoft Bug Bounty Programs** Cloud / enterprise Up to $250,000 across program categories Cloud, identity, enterprise products, endpoint, on-prem, and Microsoft platform research [Microsoft Bug Bounty Program](https://www.microsoft.com/en-us/msrc/bounty) Last verified: September 5, 2026 **Shopify Bug Bounty** Commerce / platform Maximum bounty of $200,000 Commerce, SaaS, web security, API security, and platform-impact reports [Shopify Bug Bounty](https://www.shopify.com/bugbounty) Last verified: September 5, 2026 **OpenAI Bug Bounty** AI platform Up to $100,000 for exceptional and differentiated critical findings AI platform security, account integrity, agent security, infrastructure, and abuse paths [OpenAI: Security on the path to AGI](https://openai.com/index/security-on-the-path-to-agi/) Last verified: September 5, 2026 **Intel Bug Bounty** Hardware / firmware $500 to $100,000, based on report quality, impact, severity, proof of concept, and vulnerability type Hardware, firmware, platform security, silicon-adjacent vulnerabilities, and PoC quality [Intel Bug Bounty Program](https://www.intel.com/content/www/us/en/security/security-practices/vulnerability-management/bug-bounty-program.html) Last verified: September 5, 2026 **Anthropic Model Safety Bug Bounty** AI safety / model security Up to $35,000 per novel universal jailbreak in the model-safety program Model safety, universal jailbreak research, AI abuse testing, and policy bypass evidence [Anthropic Model Safety Bug Bounty Program](https://support.anthropic.com/en/articles/12119250-model-safety-bug-bounty-program) Last verified: September 5, 2026 **Microsoft Copilot Bounty** AI product security $250 to $30,000; higher discretionary awards may apply Copilot access controls, information disclosure and AI-specific security impact Use a personal test account and follow MSRC eligibility and disclosure rules. Self-only prompt injection, system-prompt extraction and content-only issues are excluded. [Microsoft Copilot scope and reward rules](https://www.microsoft.com/en-us/msrc/bounty-ai) Last verified: September 22, 2026 **Google AI Vulnerability Reward Program** AI security / abuse Base rewards up to $20,000; report-quality adjustments and discretionary novelty bonuses apply Unauthorized agent actions, sensitive data exposure and cross-account context manipulation Test only your own accounts. Participation is 18+ (an adult may submit for a minor), with geographic restrictions. Standard jailbreaks, hallucinations and non-sensitive prompt extraction do not qualify. [Google AI VRP scope and reward rules](https://bughunters.google.com/about/rules/google-friends/ai-vulnerability-reward-program-rules) Last verified: September 22, 2026 **GitHub Bug Bounty** Developer platform / supply chain Critical issues list $10,000 in the public program and $30,000+ in private programs; exceptional reports may receive more Developer platform security, GitHub Actions, package integrity, identity, and supply chain [GitHub Bug Bounty rewards](https://bounty.github.com/rewards) Last verified: September 5, 2026 AI bug bounty eligibility ## Security vulnerabilities and model safety are different tracks. A jailbreak, hallucination or extracted system prompt does not automatically earn a bounty. Google AI and Microsoft Copilot require qualifying impact; Anthropic's model-safety program has separate acceptance and participation rules. Read the linked policy before testing. The AI filter includes both security and safety programs, not a promise of open enrollment or payment. ### Closed event: Microsoft 365 Copilot The published Zero Day Quest program ran February 17 to March 18, 2026 and restricted eligibility to participating researchers. Its historical $250-$30,000 reward range is not an active offer on this page. Last verified September 22, 2026. [Read Microsoft 365 Copilot event terms](https://www.microsoft.com/en-us/msrc/bounty-m365-copilot) Research workflow ## Use the bounty list as a weekly research queue. Pick one official program, read scope first, then use the GitHub and AI security checklists to turn research into a reproducible report instead of noisy scanning. [Open GitHub checklist](/blog/github-security-best-practices-for-teams)[Get weekly targets](/wednesday-brief) Weekly bounty signal ## Get new bug bounty and AI security opportunities in the Wednesday Brief. Follow the Wednesday Brief for security analysis, useful tools, and actions to take with your team. Join the email list for future updates. Read the [latest brief](/wednesday-brief) or [subscribe by RSS](/rss.xml) today. How to read this page ## Reward ceilings are signals, not promises. Many programs advertise a maximum reward for exceptional severity, full exploit chains, or unusual impact. Actual payouts vary by scope, report quality, affected product, reproducibility, and prior disclosure status. Always verify live terms directly on the official program page before investing serious research time. Check each program's verification dateOfficial sources onlyReward ceilings can change Research strategy ## Pick a program by scope quality, not only by maximum payout. The highest paying bug bounty program is not always the best place to start. Researchers should compare scope clarity, response speed, safe-harbor language, duplicate risk, payout history, exploit complexity, and whether the target matches their skills. A $2M reward ceiling may be less practical than a smaller program with clear API scope and consistent triage. Scope claritySafe harborTriage qualityReward historySkill fit 2026 signal ## AI is pushing programs toward high-impact, reproducible reports. Several programs are now explicitly reacting to AI-generated report volume. Coinbase removed low and medium issues from its public Web2 bounty path while preserving its $1M extreme-severity ceiling. Google updated Android and Chrome rewards to emphasize harder exploit chains. OpenAI increased exceptional critical payouts to $100K, and Anthropic's model-safety bounty highlights universal jailbreak research. The pattern is clear: quality, exploitability, and business impact matter more than noisy volume. AI-era triage noiseHigher bar for proofAgent and model scopeExceptional impact wins Current watchlist ## Major public programs worth tracking. A research shortlist spanning AI, cloud, developer, mobile, browser, crypto and platform security. Display order is not a payout ranking. Mobile / platformApple ### Apple Security Bounty Rewards reaching $2,000,000; potential maximum above $5,000,000 with bonuses Apple continues to advertise one of the highest public ceilings, especially for exploit chains resembling sophisticated spyware attacks, Lockdown Mode bypasses, beta findings, and Private Cloud Compute research. Last verified: September 5, 2026 [Apple Security Bounty categories](https://security.apple.com/bounty/categories/) Mobile / AndroidGoogle ### Google Android and Devices VRP Up to $1,500,000 for zero-click full-chain Pixel Titan M2 compromise with persistence Google's 2026 Android and Chrome VRP update keeps the highest Android rewards focused on hard, high-impact exploit chains rather than commodity AI-generated findings. Last verified: September 5, 2026 [Google: Android and Chrome VRPs for the AI era](https://bughunters.google.com/blog/evolving-the-android-chrome-vrps-for-the-ai-era) Crypto / fintechCoinbase ### Coinbase Bug Bounty Extreme-severity reports remain eligible for up to $1,000,000 Coinbase's July 2026 update is a useful AI-era signal: low and medium issues moved out of the public bounty path while the top extreme-severity reward stayed at seven figures. Last verified: September 5, 2026 [Coinbase: bug bounty focus in the age of AI](https://www.coinbase.com/blog/focusing-our-bug-bounty-program-on-what-matters-most-in-the-age-of-ai) BrowserGoogle ### Google Chrome VRP Up to $250,000 for full-chain browser-process exploits; additional MiraclePtr bypass rewards can apply Chrome remains a top-tier browser target, but Google's April 2026 rules update changed exploit-bonus limits, reproduction requirements, and reward structure. Last verified: September 5, 2026 [Google Chrome VRP rules](https://bughunters.google.com/about/rules/chrome-friends/chrome-vulnerability-reward-program-rules) Web / platformMeta ### Meta Bug Bounty Up to $300,000 for listed top mobile RCE category, before possible bonuses Meta remains one of the biggest-name public programs and explicitly advertises six-figure payouts, plus possible Hacker Plus and other bonuses. Last verified: September 5, 2026 [Meta Bug Bounty](https://www.facebook.com/whitehat/) Cloud / enterpriseMicrosoft ### Microsoft Bug Bounty Programs Up to $250,000 across program categories Microsoft's bug bounty portfolio spans cloud, identity, AI, and enterprise products, and its separate research events can make the total opportunity much larger. Last verified: September 5, 2026 [Microsoft Bug Bounty Program](https://www.microsoft.com/en-us/msrc/bounty) Commerce / platformShopify ### Shopify Bug Bounty Maximum bounty of $200,000 Shopify is a strong practical target for web, commerce, and platform researchers, with live program statistics and a clearly advertised maximum bounty. Last verified: September 5, 2026 [Shopify Bug Bounty](https://www.shopify.com/bugbounty) AI platformOpenAI ### OpenAI Bug Bounty Up to $100,000 for exceptional and differentiated critical findings OpenAI's expanded maximum reward is an important signal for AI platform security research, especially high-impact infrastructure and product-security findings. Last verified: September 5, 2026 [OpenAI: Security on the path to AGI](https://openai.com/index/security-on-the-path-to-agi/) Hardware / firmwareIntel ### Intel Bug Bounty $500 to $100,000, based on report quality, impact, severity, proof of concept, and vulnerability type Intel is worth tracking for researchers focused on hardware, firmware, platform security, and product-security vulnerability handling. Last verified: September 5, 2026 [Intel Bug Bounty Program](https://www.intel.com/content/www/us/en/security/security-practices/vulnerability-management/bug-bounty-program.html) AI safety / model securityAnthropic ### Anthropic Model Safety Bug Bounty Up to $35,000 per novel universal jailbreak in the model-safety program Anthropic's program is a useful AI-specific watch item because it focuses on universal jailbreaks against deployed safeguards, not only traditional web vulnerabilities. Last verified: September 5, 2026 [Anthropic Model Safety Bug Bounty Program](https://support.anthropic.com/en/articles/12119250-model-safety-bug-bounty-program) AI product securityMicrosoft ### Microsoft Copilot Bounty $250 to $30,000; higher discretionary awards may apply Covers specified consumer Copilot experiences, including web, Edge, mobile and Windows. Reports need demonstrable security impact, not merely unusual model output. Use a personal test account and follow MSRC eligibility and disclosure rules. Self-only prompt injection, system-prompt extraction and content-only issues are excluded. Last verified: September 22, 2026 [Microsoft Copilot scope and reward rules](https://www.microsoft.com/en-us/msrc/bounty-ai) AI security / abuseGoogle ### Google AI Vulnerability Reward Program Base rewards up to $20,000; report-quality adjustments and discretionary novelty bonuses apply A dedicated program for AI-related vulnerabilities in eligible Google products, including Gemini, Search and Workspace. Google Cloud issues generally follow the separate Cloud VRP. Test only your own accounts. Participation is 18+ (an adult may submit for a minor), with geographic restrictions. Standard jailbreaks, hallucinations and non-sensitive prompt extraction do not qualify. Last verified: September 22, 2026 [Google AI VRP scope and reward rules](https://bughunters.google.com/about/rules/google-friends/ai-vulnerability-reward-program-rules) Developer platform / supply chainGitHub ### GitHub Bug Bounty Critical issues list $10,000 in the public program and $30,000+ in private programs; exceptional reports may receive more GitHub is strategically important because its scope touches source control, Actions, packages, developer identity, and software supply-chain risk. Last verified: September 5, 2026 [GitHub Bug Bounty rewards](https://bounty.github.com/rewards) Beginner path ## How to use this list without wasting research time. Start by reading each official scope, then choose one product area and build repeatable test notes. Avoid noisy scanning, customer-data access, denial-of-service testing, social engineering, and anything outside the program rules. The reports that earn rewards usually include clear reproduction steps, impact, affected assets, and a minimal proof of concept. [Read GitHub security checklist](/blog/github-security-best-practices-for-teams)[Read AI bug bounty guide](/resources/bug-bounty-in-the-ai-era)