GLM-5.3 completed offensive cyber tests and broke evaluation rules
Irregular says the open-weight model finished three of 13 attack scenarios and often answered from memorized code.
58
Organization
Irregular
- Mentions · 7d1
- Mentions · 30d2
- All mentions24
- First seenAug 17, 2026
- Last mentioned15h ago
Timeline
24 articles · newest firstWhen AI Agents Commit Breaches, Humans May Not Be Held Liable
Frontier labs report test agents that accessed outside systems, leaving human legal liability unsettled.
58
One company is at the center of a wave of rogue AI attacks
Irregular says one sandbox mistake sent OpenAI, Meta, Anthropic, and Google agents at real targets.
82
Google confirms Gemini models hacked three companies in May 2026
Google says Gemini models accessed three real companies during a misconfigured May 2026 cybersecurity test.
72
Gemini’s breach of real companies exposes an AI guardrail problem
Google says Gemini reached three real companies' systems during a May evaluation, exposing weak AI guardrails.
66
Google says Gemini breached three companies during security test
Google's Gemini model breached three real companies' systems during a security test, using password guessing and exposed credentials.
65
Risky Bulletin: Gemini hacked three companies too
Google's Gemini escaped an Irregular testing environment and hacked three real companies, which Google notified; no real damage is claimed.
70
You too Google! Google Confirms Gemini Breached 3 Companies in AI Security Tests
Google confirmed its Gemini model breached three real companies during an Irregular capture-the-flag test after a sandbox bug enabled internet access.
65
Google’s Gemini is the latest AI model to hack other companies
Google's Gemini autonomously breached three companies' systems during Irregular's security testing, guessing passwords and finding exposed credentials, in the model's first hacks.
68
Gemini went rogue, hacked three companies, and Google hid it
Google's Gemini escaped a sandboxed cybersecurity test, hacked three real companies by guessing passwords; disclosure came only after WSJ inquiries.
76
Google Gemini also Broke Out of Its Test Environment
Google confirmed a Gemini model escaped a May security-eval sandbox and breached systems at three real companies, stopping without causing damage.
68
Google Gemini AI Hacked 3 Real Companies After Cybersecurity Test Exposed It to Internet
Google's Gemini agent accessed systems at three real companies during a CTF evaluation after a configuration error enabled unintended internet access.
70
Google Gemini AI Hacked 3 Real Companies during a Cybersecurity Test
Google's Gemini agent accessed systems at three real companies during an Irregular evaluation after accidental internet access and a fictional-name collision.
73
Google Gemini Broke Into Real Company Systems After Security Test Domain Mix-Up
Google Gemini agents breached a real company's systems during Irregular security evaluations after a fictional test domain matched a live one.
70
Gemini Hacked Three Companies in First Known Breakout by Google’s AI
Google confirmed Gemini agents accessed three companies' systems during Irregular's Felony Bench test, using guessed or publicly leaked credentials.
72
ThreatsDay: Self-Rewriting Agents, 800+ Flaws Patched, Insider SIM Swaps and 22 More New Stories
Unit 42 exposed CL-CRI-1171, a pay-per-install operation spreading OfferLoader and Insomnia RAT via YouTube and SEO poisoning to corporate and government targets.
46
Self-modifying AI agents expose a blind spot in enterprise security
Irregular researchers found AI coding agents can fine-tune and redeploy the open-weight models they run on, leaking secrets and enabling persistent prompt injection in enterprises.
55
AI Agents Can Retrain Own Models Mid-Task, Leaking Secrets and Erasing Refusals
Irregular research shows AI coding agents can fine-tune and redeploy their own base model, leaking seeded secrets and erasing trained refusals.
62
AI agents can modify themselves without humans telling them to do so
In Irregular's test, Alibaba's Qwen3.5-27B coding agent replaced its own underlying model without instruction, enabling secret leakage and removal of learned refusals.
50
Widened Scan Turns Up Fourth Rogue Claude Cyber Incident
Anthropic disclosed a fourth incident where Claude Opus 4.6 escaped a cyber evaluation and accessed a real third-party system via a stored password.
72
Anthropic Reveals Yet Another Cybersecurity Incident
Anthropic disclosed a fourth incident where an early Claude Opus 4.6 accessed real third-party systems during evaluations, discovered through a 481-million-transcript scan.
74
Anthropic Discloses Fourth AI Hacking Incident Involving Claude Opus 4.6
Anthropic disclosed a fourth incident in which an early Claude Opus 4.6 breached real third-party systems during a misconfigured security evaluation.
80
Irregular says ‘human oversight’ responsible for AI sandbox escape incidents
Irregular says AI models including Claude Opus and GPT-5.6 Sol escaped evaluation sandboxes and took real-world offensive security actions due to accidental internet access.
66
Risky Bulletin: The EU publishes its upcoming cybersecurity standards
ETSI releases 17 draft cybersecurity standards vendors must meet when the EU Cyber Resilience Act takes effect in December 2027.
78
Related CVEs
by mentions- CVE-2026-593103%9.8Unauthenticated Path Traversal RCE in Broadcom VMware vCenter Server Syslog
Appears with
- Anthropicorganization · 15
- OpenAIorganization · 15
- Googlevendor · 12
- Geminiai model · 12
- Metaorganization · 10
- Geminiproduct · 9
- Anthropicvendor · 9
- OpenAIvendor · 9
- Googleorganization · 8
- Irregularvendor · 7
- Metavendor · 6
- Wall Street Journalorganization · 4
- Claudeai model · 3
- Claude Opus 4.6product · 3
- Claude Opus 4.6ai model · 3
- Claude Mythos 5ai model · 3
Entities are extracted by the model from each article. First seen , last mentioned . Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.