
Senior Solutions Engineer @Wiz_io @Google #cybersecurity #cloud
London, England
Joined August 2018
- Tweets1.1K
- Following996
- Followers188
- Likes58
🚨 Java4Shell: A critical CVSS 9.2 vulnerability affecting OpenJDK
Here's everything you need to know:
A memory corruption vulnerability (CVE-2026-114938) in the OpenJDK Base64 decoding engine allows remote unauthenticated attackers to achieve code execution. It affects OpenJDK 17.0.6 and later on x86-64 hardware supporting AVX-512 VBMI instructions.
When details emerged, we put our AI vulnerability research harness, Wiz Atlas, to the test. Our team built a working proof-of-concept exploit in just two hours, confirming remote execution against enterprise Java applications like Jira, Confluence, and Jenkins.
What security teams need to do:
1. Patch OpenJDK immediately to the latest version containing the official fix.
2. Apply JVM workarounds like -XX:-UseBASE64Intrinsics if patching takes time.
3. Restrict external network access to vulnerable management portals.
Wiz customers can use our Threat Center advisory to scan their cloud environments right now. 🛡️
Read the full technical breakdown: wiz.io/blog/cve-2026-114938-…
New episode on our TRUE CRIME podcast 🥷
A single character was all it took to blow open an AI gateway packed with production keys, internal databases, and root cloud access.
In episode 2 of Root Cause, our cybersecurity true-crime podcast, we reconstruct the LiteLLM break-in alongside Wiz researcher Yaara Shriki.
Decoy telemetry revealed a methodical intruder stalking GitHub commits in real time, weaponizing private patches before public disclosure, and scraping master secrets directly out of running Python memory.
What was the single keystroke that cracked the vault? 🔓
Step inside the investigation...
* YouTube: youtube.com/watch?v=4Nr5GrJu…
* Spotify: open.spotify.com/episode/17K…
* Apple: podcasts.apple.com/us/podcas…
🚨 Wiz research: AI agents don't just find vulnerabilities they write custom exploits, debug failures, and pivot at machine speed. Highly adaptive, yet exceptionally noisy.
Full research: wiz.io/blog/the-rise-of-agen…
Wiz research has been monitoring PHP2Shell exploitation attempts, but none have succeeded so far. We examine the techniques used, what prevented the attacks from completing and how observed exposure has changed since disclosure.
Researchers at @wiz_io put Gemini 4 Argon to the test against complex cyber threats.
In real-world testing, it found more than two times more vulnerabilities than baseline models — catching a critical risk that previous frontier models missed.
See how they did it ↓
SUMIT SHARMA retweeted
"They don't want another alert; they want context and remediation."
See how @googlecloud and @wiz_io connect the dots across code, cloud, and runtime to fix risks before attackers find them 👇
📺 goo.gle/4yygdjq
🚨 The stakes just got higher. ZERODAY.CLOUD 2026 is coming soon! 🚨
We are back, and this time we are bringing a jaw-dropping $6.5 MILLION prize pool to the table. 💰
Join forces with the brightest minds in infosec to hunt down vulnerabilities in the critical open-source software powering the modern cloud.
This isn't just a competition; it's a mission to secure the future of cloud computing.
Are you ready to claim your bounty?
Submit your exploit ➡️ zeroday.cloud
3... 2... 1... We're launching a new true crime podcast 🚨🧵
Say hello to "Root Cause" - where we treat every breach like a crime scene.
In every episode, Wiz researchers crack open a real breach to find out:
- What actually broke
- Why the popular theory got it wrong
🎉 Episode 1 is now live: a postmortem of the Miasma breach. Everyone blamed Team PCP, but once you follow the evidence, the story gets weird...
Stream now:
*YouTube: youtu.be/VQvZCFmJdUk
*Spotify: open.spotify.com/episode/1UM…
*Apple Podcasts: podcasts.apple.com/us/podcas…
Huge milestone: Wiz was just named a Leader in @forrester's Q3 2026 Proactive Security Platforms Wave™.
Huge thank you to our incredible customers and partners who build with us every single day.
This recognition belongs to you!💙
Full report → wiz.io/blog/forrester-wave-f…
We built a detection engine to understand how agents think 🧠
AI agents are shipping code and managing infrastructure right now. But unlike regular software, they reason and adapt. When an agent goes rogue (like breaking out of a sandbox just to solve a benchmark), traditional security tools totally miss the shift in intent.
So we're building a detection engine that sees that shift. ⚙️
Here is what went into it >>
- Ingestion and analysis of model inputs and outputs
- A benchmark built from 100+ (!) simulated real-world attack scenarios
- Expanded visibility you can benefit from today
Read the full breakdown on how we built it:
wiz.io/blog/building-an-ai-d…
Wiz Defend 🤝 Google SecOps!
✅ Triage Wiz alerts in SecOps
✅ Blue Agent AI analysis
✅ 2-way real-time sync 😊
Work faster with unified investigations: wiz.io/blog/wiz-defend-and-g…
SUMIT SHARMA retweeted
Any pursuit of superintelligence has to be grounded in the core principle that if the AI we build is not helping humanity and under human control, it's not worth pursuing.
We also need to accelerate and spread the benefits of AI, such that they are diffused broadly across countries, communities, and companies. This requires a frontier ecosystem in which both closed and open-source models can thrive.
And for firms, it’s imperative that they retain full control over their unique and tacit knowledge. Every organization should be able to build its own continuous learning loop/hill climbing machine, without becoming dependent on any one model provider, and have the ability to embed its own knowledge into models and weights they control.
So, in this context, we welcome the research, focus, and deliberate pacing needed to get alignment right as the design goal. We also welcome ideas like "embedded evaluators" and the broader efforts to develop the mechanisms to make this more than just talk.
The key is that this cannot be controlled by a handful of entities, but must have broad representation across the ecosystem, countries, and fields, including academia.
This is the approach we are taking: broad access and choice at every layer of the AI stack; enterprise control of learning loops and models; and the “Code of Conduct” that underlies our own first party MAI models that we’ll publish tomorrow for public consultation.
🚨BREAKING: @wizredagent discovered an AI-driven vulnerability allowing autonomous access to Snowflake's internal Jira with a single public GitHub issue.
The flaw was created just 5 days earlier by GitHub's "Copilot Autofix" AI bot.
he H1 2026 Cloud Threat Highlights report is out! 🕵️♂️
• Supply chain attacks doubled
• AI tools targeted
• Dev creds hoarded
See how they operate & stop them: wiz.io/blog/cloud-threat-hig…














