Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

World & Business

Announcement

Anthropic expands its Cyber Verification Program for security professionals

Verified professionals can access Mythos 5.1, Opus 5.5 and Sonnet 5.5 with defensive safeguards, Anthropic says.

13 Sources, 3d ago, first seen 3d ago

TLDR

Anthropic says its expanded Cyber Verification Program gives verified security professionals broader access to Claude Mythos 5.1, Opus 5.5 and Sonnet 5.5, with safeguards designed for defensive work. New tiers also allow authorized offensive work, including penetration testing and red-teaming, though the announcement does not spell out how verification, tier access or safeguards work.

Combined views

1.2M

13 Sources, first seen 3d ago

6.4K likes479 comments2.4K saves605 reposts

Combined views

1.2M

13 Sources, first seen 3d ago

6.4K likes479 comments2.4K saves605 reposts

Anthropic is expanding its Cyber Verification Program to give verified security professionals broader access to what it calls its most capable models. The company says verified professionals can access Claude Mythos 5.1, Opus 5.5 and Sonnet 5.5 with safeguards designed for defensive work.

Featured Source

Anthropic also says it is opening new tiers for authorized offensive security work, including penetration testing and red-teaming. The announcement does not explain how verification works, what each tier allows or how the safeguards operate.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Related

Anthropic’s will ban abusive behavior toward Claude starting November 12, 2026

Usage policy update adds new rules and highlights the potential for misuse.

State of AI report puts AI-assisted AI research in focus

Author Nathan Benaich cites an Anthropic index and argues robotics is nearing a “GPT-2 moment.”

Anthropic launches Claude Haiku 5.5, its latest small AI model

Anthropic says it costs around 75% less to run on average than Haiku 4.5 and improves coding and computer use.

13 Sources

@AnthropicAIWe’re expanding our Cyber Verification Program to give security professionals broader access to our most capable models. Through this program, verified security professionals can access Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5 with safeguards designed for defensive work. We’re also opening up new tiers to allow for authorized offensive work, like penetration testing and red-teaming. https://www.anthropic.com/news/cyber-verification-program3d

@AndrewCurran_This explains the screenshots people were posting of them talking to Mythos 5.1 this morning.3d

@elder_plinius@AnthropicAI can i have railless Mythos 5.5 access plz i think it’d be good for us all to know what the real capabilities are like currently3d

@claudedailyupd🚨 Anthropic is giving vetted cybersecurity teams more access to its most capable Claude models. The expanded Cyber Verification Program will have three access tiers, with different levels of safeguards depending on the work being done. That includes access to Claude Opus 5.5, Sonnet 5.5, Mythos 5.1 and future models. The interesting part is the reason behind it. Anthropic says its earlier cyber program helped uncover 129,000 verified vulnerabilities, including 33,000 rated critical or high severity. The direction is pretty clear: instead of making powerful cyber capabilities available to everyone, Anthropic is giving trusted security researchers more room to use them. AI safety is starting to look less like “block everything dangerous” and more like “give the right people access to powerful tools under controlled conditions.”3d

@shipfrontieraiAnthropic expanded its Cyber Verification Program into three tiers, giving vetted defenders and red-teamers access to Mythos 5.1, Opus 5.5 and Sonnet 5.5 with reduced safeguards, per Anthropic. https://www.anthropic.com/news/cyber-verification-program3d

@qzAnthropic Opens AI Cyber Access to More Security Firms Anthropic has restructured its Cyber Verification Program into three tiers, giving a wider range of cybersecurity organizations access to its mos https://qz.com/anthropic-cyber-verification-program-expansion-three-tiers-100626 #CyberVerification #AnthropicClaude #CybersecurityAI3d

@AGTPinsightsAnthropic just expanded its Cyber Verification Program into three access tiers. Here's what you need to know. The program now gives verified security professionals broader access to Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5. These models come with safeguards designed for defensive work. Anthropic is also opening new tiers for authorized offensive work, like penetration testing and red-teaming. Most AI labs only let people use their models to defend, not to test by attacking. The three tiers are Defense, Red Team, and Specialized. Teams apply for the level of access that best fits their work. The higher the tier, the fewer cyber blocks. The revamped program combines two programs Anthropic has run for the past six months. The rollout is already live. Some users of the old, legacy program say its red-teaming and pentesting access seems to have been invalidated. Others say defense access alone does not allow offensive testing. Anthropic's team is urging defenders who get blocked today to apply to the Defender and Red-Team tiers. Key numbers: - 3 access tiers: Defense, Red Team, Specialized - 3 models covered: Mythos 5.1, Opus 5.5, Sonnet 5.5 - 2 earlier programs merged into one - Opus 5.5 at the Red Team tier reportedly hit 34 of 50 tasks with zero blocks - Announced Oct 6-7, 2026 The goal is to get more defenders the tools to raise the floor on software quality, with offensive work gated behind verification.3d

@rohanpaul_aiAnthropic is opening Claude Mythos 5.1 to vetted cyber defenders through 3 access tiers that absorb Project Glasswing. Until now, only Glasswing members got Mythos, while the Cyber Verification Program loosened filters on Opus and Sonnet for vetted security teams. Anthropic ran Opus 5.5 through its CyScenarioBench attack scenarios 50 times per tier, and without program access every run stopped at the first prompt. anthropic Claude now gives more hacking help the more Anthropic trusts you and the more dangerous the systems you are allowed to attack. A hospital security team or a lone researcher gets help analyzing malware and fixing flaws, but Claude won't help them attack anything. A penetration-testing firm can also have Claude break into client systems it is hired to test, though ransomware stays blocked. Only organizations vetted with the US government can use Claude to probe power grids, flight systems and interbank networks. The expansion follows Anthropic's finding that Zhipu AI's freely downloadable GLM-5.3 built working exploits in 50 of 410 attempts, nearly matching Mythos Preview's 56. Anyone can strip that model's safety refusals, as Anthropic did for about $4,400 of compute, so attackers already have near-Mythos exploit skills and so in many sense withholding Mythos mostly handicaps defenders.3d

@DanielleFongRT @AndrewCurran_: This explains the screenshots people were posting of them talking to Mythos 5.1 this morning.3d

@andresverafAnthropic folds Glasswing into 3-tier cyber access. Partners found 129,000+ verified vulns, 33,000+ critical or high. The tier for grids and flight systems is reviewed with the US government. https://www.anthropic.com/news/cyber-verification-program3d

Home
Technology
Gaming
Entertainment
World & Business
Science
Sports
AI

13 Sources

@AnthropicAIWe’re expanding our Cyber Verification Program to give security professionals broader access to our most capable models. Through this program, verified security professionals can access Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5 with safeguards designed for defensive work. We’re also opening up new tiers to allow for authorized offensive work, like penetration testing and red-teaming. https://www.anthropic.com/news/cyber-verification-program3d

@AndrewCurran_This explains the screenshots people were posting of them talking to Mythos 5.1 this morning.3d

@elder_plinius@AnthropicAI can i have railless Mythos 5.5 access plz i think it’d be good for us all to know what the real capabilities are like currently3d

@claudedailyupd🚨 Anthropic is giving vetted cybersecurity teams more access to its most capable Claude models. The expanded Cyber Verification Program will have three access tiers, with different levels of safeguards depending on the work being done. That includes access to Claude Opus 5.5, Sonnet 5.5, Mythos 5.1 and future models. The interesting part is the reason behind it. Anthropic says its earlier cyber program helped uncover 129,000 verified vulnerabilities, including 33,000 rated critical or high severity. The direction is pretty clear: instead of making powerful cyber capabilities available to everyone, Anthropic is giving trusted security researchers more room to use them. AI safety is starting to look less like “block everything dangerous” and more like “give the right people access to powerful tools under controlled conditions.”3d

@shipfrontieraiAnthropic expanded its Cyber Verification Program into three tiers, giving vetted defenders and red-teamers access to Mythos 5.1, Opus 5.5 and Sonnet 5.5 with reduced safeguards, per Anthropic. https://www.anthropic.com/news/cyber-verification-program3d

@qzAnthropic Opens AI Cyber Access to More Security Firms Anthropic has restructured its Cyber Verification Program into three tiers, giving a wider range of cybersecurity organizations access to its mos https://qz.com/anthropic-cyber-verification-program-expansion-three-tiers-100626 #CyberVerification #AnthropicClaude #CybersecurityAI3d

@AGTPinsightsAnthropic just expanded its Cyber Verification Program into three access tiers. Here's what you need to know. The program now gives verified security professionals broader access to Claude Mythos 5.1, Opus 5.5, and Sonnet 5.5. These models come with safeguards designed for defensive work. Anthropic is also opening new tiers for authorized offensive work, like penetration testing and red-teaming. Most AI labs only let people use their models to defend, not to test by attacking. The three tiers are Defense, Red Team, and Specialized. Teams apply for the level of access that best fits their work. The higher the tier, the fewer cyber blocks. The revamped program combines two programs Anthropic has run for the past six months. The rollout is already live. Some users of the old, legacy program say its red-teaming and pentesting access seems to have been invalidated. Others say defense access alone does not allow offensive testing. Anthropic's team is urging defenders who get blocked today to apply to the Defender and Red-Team tiers. Key numbers: - 3 access tiers: Defense, Red Team, Specialized - 3 models covered: Mythos 5.1, Opus 5.5, Sonnet 5.5 - 2 earlier programs merged into one - Opus 5.5 at the Red Team tier reportedly hit 34 of 50 tasks with zero blocks - Announced Oct 6-7, 2026 The goal is to get more defenders the tools to raise the floor on software quality, with offensive work gated behind verification.3d

@rohanpaul_aiAnthropic is opening Claude Mythos 5.1 to vetted cyber defenders through 3 access tiers that absorb Project Glasswing. Until now, only Glasswing members got Mythos, while the Cyber Verification Program loosened filters on Opus and Sonnet for vetted security teams. Anthropic ran Opus 5.5 through its CyScenarioBench attack scenarios 50 times per tier, and without program access every run stopped at the first prompt. anthropic Claude now gives more hacking help the more Anthropic trusts you and the more dangerous the systems you are allowed to attack. A hospital security team or a lone researcher gets help analyzing malware and fixing flaws, but Claude won't help them attack anything. A penetration-testing firm can also have Claude break into client systems it is hired to test, though ransomware stays blocked. Only organizations vetted with the US government can use Claude to probe power grids, flight systems and interbank networks. The expansion follows Anthropic's finding that Zhipu AI's freely downloadable GLM-5.3 built working exploits in 50 of 410 attempts, nearly matching Mythos Preview's 56. Anyone can strip that model's safety refusals, as Anthropic did for about $4,400 of compute, so attackers already have near-Mythos exploit skills and so in many sense withholding Mythos mostly handicaps defenders.3d

@DanielleFongRT @AndrewCurran_: This explains the screenshots people were posting of them talking to Mythos 5.1 this morning.3d

@andresverafAnthropic folds Glasswing into 3-tier cyber access. Partners found 129,000+ verified vulns, 33,000+ critical or high. The tier for grids and flight systems is reviewed with the US government. https://www.anthropic.com/news/cyber-verification-program3d

Today's Rank

—

Not ranked yet

Today's Rank

—

Not ranked yet