SafeBreach Labs found a way to manipulate Google Gemini via malicious WhatsApp, Slack, and SMS notifications.
The technique: "Fake Context Alignment"—spoofing trusted contacts, triggering unauthorized actions, poisoning long-term memory.
Read more: https://t.co/76wNZBKqMh
Turn CTEM into action.
Powered by SafeBreach Helm and AI agents, the SafeBreach CTEM Platform helps teams discover, validate, and remediate exposures—moving from reactive security to continuous risk reduction.
Learn more about SafeBreach Helm: https://t.co/XQsrAxclCG
CRINK: 4 nation-state threat actors. 4 distinct playbooks.
• China threat actors hide in your network
• Russia threat actors burn it down
• Iran threat actors hit your OT
• North Korea threat actors steal your crypto
Read the full breakdown: https://t.co/djTjS3efUM
If Volt Typhoon was already inside, would you know? Our guide breaks down how Chinese Nation-State Threat Actors like Volt Typhoon, Salt Typhoon, and APT41 quietly evade detection—and how to validate whether your defenses would stop them.
Read the guide: https://t.co/dZkLxo61nU
Static marketing battlecards are out. AI-powered competitive intelligence is in. Read our AI-First blog post, and discover how AI-first methodologies have transformed sales enablement—replacing outdated assets with a dynamic, queryable intelligence system. https://t.co/GXUyoaiPzV
New research from SafeBreach Labs: Or Yair (@oryair1999) discovered a new class of indirect prompt injection attacks against Google Gemini's voice assistant—using everyday messaging apps like WhatsApp, Slack, & SMS as the attack vector. Read the research: https://t.co/0x5SX4pK0S
No organization is off the CRINK Axis radar. China, Russia, Iran, and North Korea are targeting critical industries with stealthy, identity-driven attacks designed to evade modern defenses. Know your adversaries—read the Complete Guide to CRINK. https://t.co/VFO3ROY3Vv
New Research: Click Or Trick (CVE-2025-59199): How do you escape the Windows 11 sandbox? SafeBreach Labs uncovered that all it takes is a single user click and chaining 4 unrelated subsystems: COM, App Identity, URI quirks, and DevTools WebSockets. 🔗https://t.co/vU8r4LRg50
Heading to TyphoonCon? Catch this session on May 29 | 11:50 AM–12:30 PM. SafeBreach Labs researcher Ron Ben Yizhak will reveal how racing services for RPC endpoints led to the discovery of two Microsoft-acknowledged vulnerabilities.
See you at TyphoonCon. https://t.co/6TLgy1b8bL
When Iranian nation-state hackers are using Ethereum, the threat has moved beyond the firewall.
🎧 Listen to the podcast on how Iran's IRGC Cyber Units target organizations:
https://t.co/6UqllQjxdR
📖 Then get the Complete Guide to Iranian Threat Actors:
https://t.co/WpaSspqjzi
China, Russia, Iran, N. Korea (CRINK) are reshaping 2026 cyber risk. These nation-state threat actor groups are running at AI-driven scale with living-off-the-land tradecraft. Read the Catalog to understand the motives, methods & threat groups behind each. https://t.co/cKmaQnAeYm
Most defenders are focused on the worm.
@SafeBreach’s Adrian Culley says that’s the distraction.
“The point isn’t the worm. The point is to overwhelm defenders while the credentials walk out the back door.”
Read the Dark Reading article: https://t.co/uW3XAkzuQL
See what your EDR misses with SafeBreach Propagate.
Our enterprise-grade testing solution maps real post-breach attack paths—including lateral movement, credential theft, and privilege escalation—before attackers can exploit them. Run an attack path demo: https://t.co/PEADduMtXx
Dirty Frag (CVE-2026-43284 & CVE-2026-43500) gives attackers a near-certain path from minor foothold to full root access—silently—across virtually every major enterprise Linux distribution. Patch now. Learn more here:
🔗 https://t.co/NVQh3zkWQc
China, Russia, Iran, and North Korea (CRINK) are targeting enterprises worldwide. Our latest guide breaks down:
• Their tactics & objectives
• Industries most at risk
• What modern cyber resilience looks like
Read now:
🔗 https://t.co/M1LDpHZYWr
Why choose CTEM by SafeBreach? Because risk should be proven—not assumed.
🔹 AEV-driven validation
🔹 AI-powered orchestration
🔹 Unified risk visibility
🔹 Seamless integrations
Take control of risk. Learn more: https://t.co/VP8hQydMbB
How does a Fortune 500 energy provider prove its defenses actually work? By using SafeBreach attack simulations to continuously validate security controls, improve threat hunting, and replace assumptions with measurable proof. 👉 Read the case study: https://t.co/DBhbFaT6tc
Heading to the Gartner Summit? Stop by our booth to see how SafeBreach Helm, our AI CTEM Agent, helps teams identify and validate exposures, trigger validations, analyze attack paths, and accelerate remediation—all from one unified platform. 🔗Learn more: https://t.co/zbb4bBLVpU