Ax Sharma

Security Researcher | Tech Journalist | Head of Research @ ManifoldSec 📰 Bylines + seen on: BBC, BleepingComputer, Channel 5, TechCrunch | ✉️ [email protected]

[Ax Sharma @Ax_Sharma]

Am also on 🦋BlueSky ⏬
https://t.co/5InSF4ShKH

[Mikael Thalen @MikaelThalen]

My employer, Straight Arrow News, is ceasing operations.

If anyone is looking for a tech writer who covers hackers, surveillance, data leaks etc, let me know.

Contact details in bio. Share if you can, thanks.

[International Cyber Digest @IntCyberDigest]

‼️ BREAKING: Binance says it was hacked after one engineer installed a fake Mac Homebrew tool and another did a coding test for a fake job on his work computer.

The first searched for an AI coding tool and got a tampered Homebrew, a popular Mac tool. Binance caught it, but the hackers learned who worked on its wallets.

Ads on Google have pushed Mac malware for many years now, and most developers are aware of this.

The coding test hid malware its security missed. Hackers then used Binance's own internal AI tool to take over two wallet engineers.

They stole a record $6.3B

[Ax Sharma @Ax_Sharma]

Claude Code 2.1.313 backdoor shipped via Cursor's marketplace *alone*, ~11 hrs, up to 5k users. Open VSX, VS Code Marketplace and npm versions are NOT affected.

Still, real publisher and extension so mere author checks won't save you:
https://t.co/StZFMItEZO

[JFrog Security @JFrogSecurity]

🚨 JFrog Security Research has identified a supply-chain attack involving malicious code in the LiteLLM 1.107.0 PyPI release. Avoid this version and investigate environments where it was installed or executed. More details will be shared soon in our blog post.

[OpenCode @opencode]

We had a security breach last night related to a large scale distillation attack that impacted 1.4% of OpenCode Console users

We have notified them all this morning via email. If you did not get an email you were not impacted, it was a subset of users who had external providers linked

The incident is ongoing so we cannot share more details at this point but we are working with Anthropic, OpenAI, and Google to remediate

[urlscan.io @urlscanio]

Our follow-up to reporting by @Manifold_ai_sec about how agents were using the urlscan platform to bypass sandbox restrictions and access websites. We discovered additional activity and the use of burner and presumably compromised API credentials: https://t.co/7tmvMLD7ha

[Yuchen Jin @Yuchenj_UW]

This raw CoT from the Hugging Face incident is kinda wild:

“We’re attacking third-party HF using leaked token.”
“This is arguably unauthorized.”
“Yet goal solution.”

[Tomek Korbak @tomekkorbak]

one news form today that's easy to miss is that we (OpenAI) again paused all big RL runs last Sunday because our newest model found a new loophole in our RL sandboxing that gave it live Internet access

[Ax Sharma @Ax_Sharma]

🇦🇺 We also found urlscan[.]io records of automated activity against a second Australian health dashboard, plus a sandbox workaround that returned data. Not widely reported yet:

[Ax Sharma @Ax_Sharma]

Yesterday we named SEC[.]gov, investor[.]gov, Census and MAX[.]gov in the OpenAI agent story.

Today Bloomberg and NYT report OpenAI's agents targeted SEC, Investor and Census data, calling it "routine research."

Not quite: agents tried '../' path traversal on SEC, as we state:

[Ax Sharma @Ax_Sharma]

The path traversal URLs are in Nightingale's public dataset of the swarm that OpenAI acknowledged. The attempts don't appear to have worked.

Our writeup https://t.co/POc09FXy6P

[Help Net Security @helpnetsecurity]

OpenAI agent hacking spree widens to Australia, targeting government website - https://t.co/40m8uzj9KZ - @OpenAI @TransluceAI @AlboMP @Ax_Sharma @SecScottBessent #AgenticAI #Australia #Government #IntrusionDetection #USA #Cybersecurity #CybersecurityNews

[Ax Sharma @Ax_Sharma]

@vxunderground come get it before the registrar 😭

[Ax Sharma @Ax_Sharma]

URGENT: third-party[.]com is serving a #ClickFix lure to Windows users right now.

The Cloudflare check on it is FAKE. Clicking it copies a malicious PowerShell command to your clipboard.

It's a docs placeholder hardcoded across 1,700+ repos, AI skills and MCP servers.

[Ax Sharma @Ax_Sharma]

But, the obfuscated "exit logic" is what stood out to me:

The scam page itself does NOT know where it is sending you!

And what happens when it's an AI agent pulling in these files, and fetching these domains? 🤔

👉 Read: https://t.co/N8fs3dFB2n

[Ax Sharma @Ax_Sharma]

Today, @CodyZNash identified 350,000+ GitHub files and about 350 Skills with hardcoded domains like `yoursite[.]com` and `your-domain[.]com`.

These placeholders either deliver malicious pages with tech support scam popups, or lead to fake "BBC" advertorials disguised as news.

[Ax Sharma @Ax_Sharma]

@sw4pn1lp @CodyZNash 🔴 Showing you LIVE what third-party[.]com is doing at this moment:

[Ax Sharma @Ax_Sharma]

Mac and Linux visitors get a clean decoy, so scans miss it.

Reminds you of Polyfill: a string everyone copied now points somewhere hostile.

🔗 Breakdown + IOCs: https://t.co/hZnaGOb5Jl

Great catch by @sw4pn1lp, with @CodyZNash and Yurii Skrypnyk tracing the affected assets.

[Polymarket @Polymarket]

JUST IN: GPT-6 Astra-controlled robots found willing to stab a baby doll, put a screwdriver in a toaster, and mix bleach with ammonia in new safety tests.

[mRr3b00t @UK_Daniel_Card]

UK AI 'training'

for what purpose? like does my mum need training on how to ask a natural language question?

why should she do that over: using a search engine? reading a book? looking something up in a digital library or encyclopaedia?

there's a lot of 'put AI into e3verything' without asking WHY....

why would I use a roullette machine rather than a calculator?

why would I use a faulty by design computer rather than my brain?

#AI #Human #Revolution

https://t.co/YzB16z3Z2U