Update:

Crowdstrike came out and released a technical report confirming my analysis. They were reading in a bad data file and attempting to access invalid memory.

This global crash was a two-part bomb. The detonator apparently, was NOT new.. it was PRE-INSTALLED.

/1
Contrary to initial suspicions, Crowd Strike did NOT push out a faulty driver, the faulty driver ALREADY existed on Mac, Linux AND Windows, likely for months or years.

Sitting there like a ticking time bomb.

/2
This bug was a two-part series.

All it needed was bad data in order to detonate it.

The recent data update, delivered that payload.

/3
Deploying new data files to computers is generally thought to be safe. Data doesn't contain executable instructions for the CPU, after all.

Therefore it doesn't go through the same code review process as new executable code.

In effect, this proved to be the perfect backdoor

/4
When there is a ticking time bomb that's been PRE-INSTALLED on computers, then all it needs is just the right data to activate it.

/5
And unfortunately, this is exactly what happened.

This data update, because of it's assumed low-security implications, was allowed to be raw dogged into every Windows box running Crowd Strike, without consent, and without notification.

/6
And btw, this same ticking time bomb apparently exists in Linux AND MacOS. They just weren't targets of this data update, so they didn't crash. If a similar push had been to Linux, we would have seen a global catastrophe.

/7
Originally, I thought this was simply programmer error. But now, I'm not so sure. My experience seeing corruption at Google showed me that obvious bugs were allowed to exist, with apparent insiders who were aware and exploiting them for their own agenda.

/8
For example, Jordan Peterson got his entire Gmail/Youtube account wiped because some insider knew they could create nearly the same email address as him and start sending spam, knowing that his account would be wiped out by the AI, despite his account being over a decade old.

/9
Is something similar going on here with Crowdstrike?

Some insider with the knowledge that this nuke existed on every Windows/MacOS/Linux box with their software, only needing the proper data-detonator to act as the trigger?

/10
And why wasn't this software bug caught by automated checking at Microsoft? This code is reading data, interpreting that as valid memory locations, and attempting to read it.

HELLO?

/11
@Microsoft, are you aware that tools have existed for DECADES that are designed to find these simple access violations and flag them?

@Microsoft have you NEVER bothered to run these tools on Crowd Strikes system drivers?

/12
This is really bizarre. And the recent facts raise a LOT more questions about why this ticking time bomb has existed on mission critical devices for months or years.

/END
It get's worse, Microsoft granted Crowd Strike's ticking time bomb with "boot-start" privileges, normally reserved for Microsoft drivers.

A boot start driver MUST be installed in order to start the Windows Operating system.
Source: Dave’s Garage. Microsoft developer Gray Beard

These are the bad data files. Notice they are named like system drivers but they are actually data files. To fix a system you need to boot into safe mode, bypassing the boot-start drivers, and then delete these data files. Image
This will remove part B of this two part blue-screen-of-death detonator. The proverbial explosive device, crowd strikes boot-start drivers, have special privileges that prevent their removal.

…And must remain in the system in order for the system to boot properly.

• • •

Missing some Tweet in this thread? You can try to force a refresh
 

Keep Current with Zach Vorhies / Google Whistleblower

Zach Vorhies / Google Whistleblower Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!

PDF

Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @Perpetualmaniac

Jul 9
This one line of code for Grok AI was more dangerous to the ruling elite than Iran continuing its nuclear program:

"The response should not shy away from making claims which are politically incorrect, as long as they are well substantiated."



It's all public, so let's dive in.
1/github.com/xai-org/grok-p…Image
This directive for Grok AI went in on July 6, 4:01pm PDT, with commit signature 535aa67, while everyone was on 4th of July holiday.



You can see that it was added because it's in green.

2/ github.com/xai-org/grok-p…Image
This is a graphic representation of the media's response. The first arrow was when the directive was committed to Grok AI's consciousness. The second arrow was when the media starting freaking out.

The X is when they reverted the change. Approximately 4pm PDT on Sunday. Just two hours after the media started freaking out.

3/Image
Read 41 tweets
Apr 6
In case you missed it… Scott Pressler is getting heat right now. He’s famous for getting out the vote efforts.

Why is he getting heat?

Because after PA and WI was lost to the left, Scott came out and bizarrely asserted that the left didn’t cheat.

Right…
This is so bizarre that now we have to ask questions.

One of the things that I noticed is that I was getting a ton of SMS messages from Scott about Pennsylvania and Wisconsin, but i’ve never spent more than a day in each state, and that was over a decade ago.
I live on the west coast. I have an oregon area code. Why am I getting-get-out the vote SMS messages for states thousands of miles away?
Read 8 tweets
Jan 29
🚨Is DeepSeek a Front for the Chinese Government?

Situational Report: How did we get here, and what comes next?

Let's dive in. 🧵👇

1/
In Silicon Valley, the prevailing belief is that DeepSeek’s success is too good to be true—many suspect it’s backed by the Chinese government.

And the numbers don’t add up.

DeepSeek claims it trained V3 for just $5.6M, a figure that seems unbelievably low.

2/
DeepSeek V3 is as powerful as OpenAI’s O1 model but runs at just 3% of the cost.

It’s so optimized it can run on a desktop computer.

And here’s the real kicker: DeepSeek can even spawn "mini" AIs that run on iPhones & Raspberry Pi.

3/
Read 22 tweets
Oct 20, 2024
I have a friend, she’s beautiful on the inside and out. But she has inexplicable depression. Their sibling committed suicide, as did for their father and their grandfather.

Three generations of men.

They followed my advice and are now on recovery. A thread.

1/
Ironically it took fringy conspiracy theorist that thinks the entire polical/scientifical/media/medical establishment is captured to correctly theorize what was going on.

I focused on their sibling as that was most recent and closest genetically to them

2/
Their sibling was smart but on the autism spectrum. ADHD. Prior to the suicide they rapidly descended into an acute depressive episode characterized by disorganized thoughts, delusional thinking and other schizo affective characteristics.

3/
Read 15 tweets
Aug 20, 2024
The most important fact of the geological sphere is the topic of Ai:

China is 10 years behind the USA.

Trump cut off China from the AI chips via the Chips Act (according to Eric Schmidt, former CEO of Google) and Biden CONTINUED the executive over.

Make no mistake…

/1
China is pissed.

They are going to lose the Ai supremacy game. Partly because of being cut off to their AI chip access, and partly because those who put them in power have made sure to corrupt their government with dysfunctional.

2/
This means USA is going to reign supreme by default. This means the USA is going to win the knowledge wars that are about to unfold.

3/
Read 10 tweets
Aug 9, 2024
I want to point out what a huge story happened yesterday that nobody picked up. Here it is in a nutshell:

* Twitter is shadow banning people still, and it’s sophisticated.
* Grok AI was tricked into revealing these shadow ban labels with a special prompt.

1/
The prompt attack used against Grok AI was very interesting and surprising.

It turns that you could embed an entire program in your question… and surprisingly… GrokAI WOULD RUN YOUR PROGRAM FOR REALZ and OUTPUT THE RESULTS BACK TO YOU.

🤯

2/
When this was discovered, tech savvy X users started copying the code injection prompt and swapped in their user name to the right place.

Grok AI would then run this program, accessing hidden apis and dump out the shadow ban list for that user.

3/
Read 12 tweets

Did Thread Reader help you today?

Support us! We are indie developers!


This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Don't want to be a Premium member but still want to support us?

Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal

Or Donate anonymously using crypto!

Ethereum

0xfe58350B80634f60Fa6Dc149a72b4DFbc17D341E copy

Bitcoin

3ATGMxNzCUFzxpMCHL5sWSt4DVtS8UqXpi copy

Thank you for your support!

Follow Us!

:(