Geek of the House
Helping the world, one n00b at a time...
Sunday, December 7, 2025
AI and reality distortion: Perspective on the progress of AI in Entertainment and Mainstream Media + Dark AI
Tuesday, April 22, 2025
A chat about #CyberHygiene
I talk about this in my #AwarenessTrainings that I put together for my company. People sometimes have the mistaken notion that they aren't targets for bad actors because they aren't famous and don't have a high net worth, or don't have a high-profile job. But that's simply not the case today. Anyone with any online presence is a potential target to attackers. That means everyone needs to know their cyber hygiene. So what does that look like?
Basic cyber hygiene is essential and easy. Steps include (extra details below):
➡️ Be more stringent about the info you share online π
➡️ Review and adjust #privacy settings π
➡️ Use strong and unique #passwords π️
➡️ Enable two-factor #authentication π️
➡️ #Monitor online presence π
➡️ Learn about data brokers ⬅️
➡️ Secure all devices π
➡️ Be skeptical of unsolicited requests π―
➡️ Regularly audit third-party apps with access to your accounts ❗
➡️ Monitor credit reports π°
➡️ Separate personal and professional identities π¬
π Sharing online: Especially if your posts online are public, be aware that anyone, including the bad guys, will see the post, its likes, and its comments unfiltered. They can use details and media included in social engineering attacks against you and your connections. They can clone your voice, or your persona, from a short video and then replay that to your connections to defraud them, or you. They can "guess" your passwords and/or security question answers just by browsing your social posts and comments. This is called #SocialEngineering
πPrivacy settings: Make sure to regularly review these settings on all sites and portals you frequent. Providers regularly update settings and provide new functionalities.
π️ Passwords & MFA/2FA: It's 2025 - you should be using a formal password manager that can provide long, strong, and unique passwords for every site/service you use. Some of them also provide OTP and QR code scanning capabilities for MFA/2FA. 1Password, Bitwarden, and Proton Pass are solid options in this space, among others. Make sure to fully vet whomever you choose (ensure they have a solid internal security policy and zero-knowledge framework at a minimum). Otherwise, use a 3rd party app like #GoogleAuthenticator or #MicrosoftAuthenticator to store your OTPs + a password manager. ProtonPass has a free tier that is very good, I use it myself. We use 1Password at my company, everyone loves it lol. Regardless of what tool you use, BE SURE TO USE ONE! The built-in browser password managers are not as secure, and only work with websites - they are basically fancy auto-fill tools, not a proper credential manager. A formal password manager will come with apps that work across all devices and operating systems so you can easily access your secrets anywhere you are.
π Monitor yourself: You should be aware of your #DigitalFootprint. Web search yourself, your email addresses, your phone numbers - find out what's out there about you. Google Alerts is a service where you can setup monitors for most any "topic." I've used them for years to search my name, email address, and the same for my immediate family. Whenever the "topic" pops up on a web page, I get an email alert. Also, you should be aware of any #BreachData out there with your info. I suggest https://haveibeenpwnd.com as a great starting point.
⬅️ Data brokers are horrible banes of our digital existence. They collect our data from so many places, most without our direct knowledge, and resell it for profit. Then they get hacked and our data is leaked outside of our control. Learn about these companies. Find ways to request for your data to be deleted from their platform. It may not be easy to erase this data, but persistence can be a good thing in most cases. Don't be surprised if you get stuck in a loop or outright denied when dealing with these companies. Your data is their profit - the don't want to give it up without a fight.
π Securing all of your devices sounds like a no-brainer, right? Wrong. How many parents leave their devices unprotected so they can permit their kids to play a game or watch a show/movie? I also see people using weak security methods to secure their devices - simple PIN codes that are 11111 or 12345 or MATCHING YOUR ATM PIN π (please don't do that!). Using a #biometric method (fingerprint or face or eye) is the most secure. Your device should also make you create a PIN as a backup to biometric - please use something good! I personally use a PIN with 6+ digits. Makes it much more difficult to brute-force.
π― Be skeptical of random DMs and emails, and especially of things that are "too good to be true". Spoiler alert! They usually are not good, nor true. Many hackers will use these methods to social engineer you. #Phishing is still the most common method of compromise globally. Email servers block trillions of emails per day as SPAM, most never even make it to your mailboxes Junk folder, but with AI the scammers are getting past even the most modern security platforms. #BeCyberAware #StayVigilant
❗Audit 3rd party app access to your accounts regularly. Remember that game you stopped playing a few months ago? You deleted the app, so you're good...right? Not if you signed into the app with your a social or email account. They still have access to your profile data and whatever else you gave them permission to see when you accepted those terms of service when you first opened the app! Make sure to go into your account's Security section and see what apps and services are still connected to your accounts. You should immediately delete/revoke any apps/services you no longer use, or you don't want to have connected to your profiles. This can include the aforementioned...Data Brokers π»π±
π°Monitor your credit reports! You want to check for any new credit or bank accounts you didn't apply for yourself, activity you don't recognize, loans taken out in your name - anything "weird or unusual". In the US, the Big 3 are: TransUnion, Experian, and Equifax. They all have free account levels, but you can of course pay for premium services if you want.
π¬Separating your personal and professional identities is important. Stop using your work computer and email for personal shit (please)! I can almost guarantee that is against [insert company name here]'s policy! As a security guy, I can tell you we really don't need to see your tax return, resume, pet's health records, your sister's/brother's picture from that [insert family event here] you went to last weekend, and I certainly don't want to see your shady Internet activity (seriously, please stop that shit). You don't want your company in your personal stuff, and your company doesn't want their stuff on mixed with your personal stuff. YOU HAVE A SMARTPHONE! Use that instead (please). If you need a PC for home/travel use, go buy a cheap one from the local electronics store. You can get a decent machine for ~$200 today that will handle simple web browsing, checking emails, and even voice or video chats. You can even connect your earbuds or headset to them. Go watch porn on your own time bruh! π
Now you're probably wondering what AI has to do with all of this. Well, all modern AI's (as of this blog post) are trained off of Internet data, along with many other training databases. So if you've ever put it out there on the Internet, ever (this blog included), it is being aggregated and processed by multiple AI foundation models and used to respond to user queries. This includes pictures and videos you have uploaded pretty much anywhere. Keep in mind there are a few #DarkAI models out there now as well that have zero guardrails, meaning anything is available unfiltered. All you need is a creative prompt.
It's a brave new world in 2025 ya'll. You need to know what is out there about you - because the hackers already do too.
Monday, July 22, 2024
#CrowdStrike Cause a Global Tech Outage - what happened, why, and (how) can it be prevented?
Sunday, August 20, 2023
"Hackers are good. Infosec is evil."
Tuesday, December 20, 2022
Thursday, May 30, 2019
Crackers and AI - A scary cool future, happening now
Crackers. They are everywhere, and better than you can imagine. They're smarter than you, me, and even those geeky people down the hall with a dozen more certs than I have or could want, and 15 years longer in the game - combined. Worst part is, there's no way to stop them anymore, unless you shut off the internet, which we all know isn't going to happen. I miss the days when installing a firewall and Antivirus was all you needed to keep them at bay. Things were much simpler then.
Today, it's a completely different game. And it scares me to be honest. They're using AI and neural networks to power their cause to the nth level. These systems aggregate data from everywhere, and I mean everywhere - clearnet and deepweb - especially social media, and then building highly sophisticated malware campaigns, that use this massive trove of data to bounce from continent to continent, changing IP ranges faster than we can block them, with techniques that aren't detected by even advanced security systems, faster than we can comprehend what's actually happening. Using real-time language translation, following local colloquialisms (on the fly) for their phishing campaigns, across any platform, in ways that are indecernable from a human doing the same thing, just better, faster, and constantly. By the way, this process happens in minutes - not days, weeks, or even months or years.
We humanz are feeding these AI's willingly, and have been for decades now. Unfortunately, it's already too late to change the situation. If you're reading this article, you're part of the problem too, just like me. The best we can do is minimize our exposure, and control what data is put on the web moving forward. In fact, it's all we can do. I liken it to sitting on a lit stick of dynamite but you can't see the fuse. You know it will explode, badly, but never when. And when it does finally pop, you quickly realize that was just the distraction. There's so much more on the back side of that. Think iceberg meets Titanic...we're on the top, barely staying above the water line and think "oh that's it" when what's really happening is multitudes worse. Let that sink in for a second. Grasp that concept and embrace it. Thats the cyber wilderness today.
As humanz, we're only capable of doing so much in our day. We need rest and fuel to function at our best. We spend months or years training and retraining, and educating ourselves, inventing new systems/processes to enhance our general experiences. Problem is, criminals don't care about any of that. They already know our next ten moves. We find a way to block them? They find a zero-day. Plus, Computers don't have human restrictions. As long as the power is on, they're always on. They don't see "time", that is a human construct. All a system sees is 1's and 0's. To these incredibly intelligent systems, we're just another node on the net. Another tool for them to expand their botnets and proliferate their cause, which is chaos, regardless of the source code. Crackers build the source code, plug its methods into a construct, and let it go. It learns exponentially, without further human interaction. It functions at the speed of light. It carries out actions that we see as malicious intent. To these machines, it's just another line of code. They don't see the emotional impact of taking over a user's machine, deleting all of its data in place of its own code or harvesting login credentials, and then using this new zombie as a stage to proliferate its cause creating more zombies, simply following it's source code. Achieving its purpose, whatever that may be.
So the question that comes up now is how does one minimize their exposure? Outside of a complete disconnect, which isn't 100% possible anymore with the amount of surveillance equipment running globally and the fact that the person sitting next to you has a smart phone that is listening (yes, they all listen), one of the best methods is to delete your social media accounts. Another compliment to that is to unsubscribe from every newsletter or website that isn't necessary for you to live, which cleans up your inboxes. Now for many, deleting social media is simply not an option. I get it. You get an endorphin hit with every 'Like' or comment on your posts. That's how it's designed, that's not your fault. You get a rush with every argument you start online. Trust me, I get the psychological need for these activities to occur. It's not me, but I still get it. So, for those of you not willing to delete your social media accounts, how can you minimize your exposure? Here's some tips. Be extra mindful about what you post to any online medium: Facebook, Twitter, Tumblr, Reddit, Instagram, LinkedIn, deep web forums, etc. Anything you post publicly will be aggregated by these AI's. Pictures (from which They can develop a facial recognition data from), speech patterns in comments (from which They can develop develop language recognition and translation data from, including local collquialisms), friends posts (which can be used for Them to make social connections, providing additional attack vectors and 'friendly' associations with which to use against you, and anyone else connected to you and your connections), and so on. Just like law enforcement recommends to not post on Facebook that you're going on vacation so thieves know when to rob your home, the same concept goes with being more secure online in general. If you wouldn't tell a stranger 'something' in real life, or wouldn't share a picture with your boss in real life (for example), don't do it online. One of the differences between real life and the internet is once you post it on the internet, it's there forever. Your boss is likely to forget 'something' you told them or shared with them in a few minutes. Computers never forget, even when you do. Remember that.
Another tip is don't reuse passwords across sites. That way if a site is compromised, the criminals can't exploit your other accounts with credential reuse (one of the most common attacks) and you don't have to change every password for every account you own (because they all used the same password). Get yourself a password manager, whether online or offline (I prefer the later personally), point is use one. There are plenty of options out there, the most popular ones I have seen in use are DashLane, LastPass, myki, and 1Password. I personally use KeepassX which is a 100% offline solution. That's what works for me, do what's best for you. The point is you need to use complex passwords that are different for every site and service you use.
Another tip is to setup 2FA/MFA for every account you can, especially bank accounts and email accounts. This gives you an extra layer of security in that not only do you need a password, but also a random code to authenticate into a given site/service. You can use an app like Google Authenticator, or Microsoft Authenticator, to store your 2FA/MFA tokens. That way, even if someone were able to figure out your password, they would also have to physically have your phone in hand in order to get in to your account. I recommend using an authenticator app versus SMS verifications because of the well know (but hard to exploit) SS7 vulnerability inherit to all mobile networks globally.
The SS7 vulnerability is well documented, you can read articles about what that is and how it works with some simple Google searches, so I won't go into what that is here.
My final tips for minimizing your exposure to the super intelligent AI's is use nicknames where ever possible, and don't use your actual picture (showing your face) as your profile picture on any site anywhere.
Now that I've completely scared you (and I hope I have) this is not to say that all AI is bad. In fact quite the contrary is true. AI is empowering our future, here and now. I use it in my professional life to help thwart cyber threats, to make my colleagues more efficient at their jobs, and to intelligently route phone calls through my call center. Many of us are touched every moment of every day by helpful AI. Power companies use AI to not only deliver services to us, but also to predict outages, etc. Siri, Cortana, Bixbi, and Google Assistant are all very helpful and non-malicious AI constructs that most of us use all the time. Scientists are using AI to develop cures for cancers. Banks use AI to provide us with loan approvals in seconds, and protect our accounts from criminals. The list truly does go on with the positives that AI brings to our modern world.
My point of telling you this and scaring the shit out of you? Awareness. Most of the connected world is still very much oblivious to what is really going on and how their online interactions affect everyone else who is connected. In my career I watch the good, bad, and ugly of our digital world unfold at a rate of billions of bytes per second. I watch these newly born AI constructs attack and protect at the same time. And while the nerd in me loves the digital world we live in and are evolving into, the Geek in me is always playing devils advocate asking why and how. Ever questioning for the real answer.
This is the world we live in today, and it will only continue to evolve in this way as time goes on. Next we'll hear about an AI being dropped into a quantum computer...and there's literally no telling what that system will decide once it realizes it exists. There's plenty of theories and conjectures, but no one really knows until we get there.
Stay safe and #CyberAware everyone. As always, feel free to comment if you have questions or want to discuss another point in greater detail. Just be mindful of what you post. They're watching us, after all 😁
Sunday, July 2, 2017
Time to Air Gap
In a world of 24 hour activity literally being streamed in real time across the globe and beyond at a rate of trillions of bytes per second at the speed of light, the biggest question in privacy is how to achieve anonymity in a world where almost nothing is secret. There is still a way to go 99%. It's simpler than you think, but an effort nonetheless.
Some background. Everything we do in digital form is cataloged and stored in a vast array of databases and servers across an amazing amount of touch points, which is then synchronized across a thousand other servers for redundancy and caching, which is then backed up to dozens of other servers, with their own redundant backups. Anything you put online...any application you use...any "terms of service" you agree to...any text or media you post...remains online forever. With the right tools and search terms, anything can be searched for, or spyed on, or downloaded in an instant. It's been this way for decades, and will continue to be that way for centuries to come, especially with as connected as the planet is and as long as there is electricity.
Some discussion. Cyber attacks are a constant thing. Increasingly, we should take as a starting point that cybersecurity compromises are the third certainty in life. The cyber world is constantly at war with itself. Governments hacking governments. Corporations hacking corporations. Governments hacking corporations. Hackers hacking governments and corporations. Hackers hacking hackers. Governments and corporations hacking hackers. And then there's everyone else. Generally oblivious. Privacy is a luxury, which we give up willingly every single second of every day. The emergence of intelligent systems, artifical neural networks, and deep thinking algorithms only proliferate this further. They take, store, and learn from every bit of data we leave as breadcrumbs. Artificial intelligence is here, and it is learning. From us. And we're letting it. Give it enough processing power, and it becomes self aware. Quantum computers will make that very real, very soon.
Some perspective. Having lived through the evolution of modern computing, including the Internet, all of this is absolutely fucking amazing, and a geek's ultimate wet dream. A demonstration of true humanz genius, ingenuity, and progress (not as far as we should be, but progress nonetheless). Highly impressive in the vastness of its brilliance and simple complexity. I Iove using It, and learning about It, and protecting It. All of it, if I am completely honest, scares the living shit out of me. There is too much. It has become frightening. AI is now making decisions and inferences faster than humans, and has even been seen generating its own programming code. So, the concept of air gapping entered my mind as a way to keep safer than I already am. Most cannot see the signs, or do not want to admit they exist, however I am of the firm belief that World War III has been well underway, and we need to protect ourselves, especially our digital lives as I feel they are the most vulnerable to compromise. Stay with me, it's all relevant.
It has been discussed for decades that the next major global war would be fought half online, and half in the real world. The evidence is all there, and I do not believe it to be simple coincidence. Global newz outlets, small town newz papers, radio ztations, and zocial media have been propagating images of this war. Pick a topic...WMD's, genocide, terrorism, ransomware, deep web market hackz and seizures, arrests of crackers and phreakz, data breaches, RFID implants, cyber surveillance initiatives, counter cyber terrorism, weapons trafficking, the unavailability of bullets to the public, gun control politics, powerful botnetz, election hacks, political hackz, hardened/weaponized computer systems...I hope you get the point.
Back full circle. Traditionally, air gapping a system means it doesn't have any network interface cards, or external drives with which to access or extract the data contained within said system. You can not get close enough to implant a listening device that reads vibrations or thermal changes being given off by the system's internal hardware to convert that into bits representing the data being actively accessed (such as login credentials, encryption/decryption key exchanges, data manipulations, etc.). The only way to extract the data contained within is by sitting at the console and physically removing the locked and encrypted drives, if there is no SD port. Then, if you can pull the impossible off (which includes getting the data off campus), you would need supercomputer power to decrypt the contents of the drive, which would still take 1,000 years to break (if and unless you are lucky). There is still the idea that once you decrypt the data, it could transmit its location to its owner, meaning you too would need an air gapped system to exfilitrate the data. Then comes what you do with said data. Yet another catch 22. The NSA, CIA, FBI, DEA, DHS, militaries, every government, and super corporations maintain their most secret data on air gapped systems. Physical access to these systems is extremely limited and highly controlled. It's considered the safest digital platform because the system isn't connected to anything but a power cable, and thus, in theory, cannot be hacked. A true digital safe, as it were. We know anything can be hacked, it just takes time. As a hacker, we count on human error and complacency, making even air gapping a 99% solution, and the best we've got. Now, take this concept and apply it to a human life. It's far simpler, and also 99%. The anomaly, is human nature.
Based on my research, here is what I have learned about how to go 99% off grid digitally. While I do not yet practice everything I note here, I am closer than even those who know me best are even aware.
1) Get off the internet, period. No social media, no surfing the clearnet, no online purchases, no clearnet email accounts. If it becomes absolutely necessary to access the Internet for a specific purpose, there are completely anonymous ways to do these tasks, on secure systems like Tails over TOR, for example, using cryptocurrency, and ghost mailboxes. Avoid Google at all costs. Use TOR browser, responsibly (www.torproject.org). But generally, just leave it all. Stop posting immediately, delete your accounts, and never go back.
2) Get rid of your smartphones, tablets, Windows and Apple computers, smart devices (TV's and refrigerators included), iRobots, etc. Need a cell phone? Buy a prepaid flip phone, and change it (and the number) every month (aka burners). Every phone can eventually be traced and tracked. Still need a computer? Learn Linux, how to secure it, and practice way smarter browsing habits (use TOR browser), if you browse at all. Keep in touch with world events, anonymously, and continuously hone your skills.
3) Always use cash or cryptocurrency, for everything. If you have to make an online purchase, use cryptocurrency, the deep web (local markets only, don't buy overseas, and be very careful), and have it shipped somewhere that is not your home, like a post office box, a business, or an associate's location, under a false name. By the way, there are ATM's now that you can convert cash to BTC, and visa versa. Look it up using Duckduckgo.com (a safe search engine).
4) Drive an older car that does not have a computer in it, or at least has all analog systems. Yes, cars are also being hacked, remotely. Keep it clean and running well though, you don't want to draw undue attention. Walk or take public transportation when you can, avoiding direct face contact with cameras. When you do go places, change your entry/exit routes regularly...avoid habitual patterns, unless necessary to remain hidden in plain site (like going to work, or getting groceries).
5) If you must, have an immaculate and purposeful digital/public footprint. Which means a clean record, and a "normal" looking life, so as to not draw undue attention. Keep it super minimal and protected, even fake some details if you wish, but it has to be believable. Your outward personality must seem conforming, friendly, and genuine. When people search for you online, they need to find only what you want them to find. Purposeful is the key word here. To keep your accounts secure, use a dice word list to generate passphrases with an entropy of 7 to 10 or more words (as the host allows), and rotate passwords on a schedule.
6) Second most important after getting offline, and the best to mention as the final advice, would be live simple and minimalistic. Only get what you literally need to live comfortably, and look "normal". The trick about hiding in plain site is being distant enough that people respect your privacy, but involved enough that they believe you to be a "normal, nice guy/gal". Avoid run-ins with the law and reporters. Do not have public arguments. Remain intelligent, articulate, empathetic, determined, and most of all inquisitive. Question anything, be aware of everything.
If you can literally get out of dodge and move to the mountains in the middle of nowhere, or something like that, the closer to 99% you get. If you are not online, there is nothing to take/attack. Here again, human nature is the anomaly.
You can be connected, yet a ghost. You can see the world, without a face. You can reach out, without being reachable. The less connected you can maintain, the better. I am committed. How far are you willing to go?
~Geek
This blog is only to express the opinions of the creator. Inline tags above link to external sites to further your understanding of current methods and/or technologies in use, or to clarify meaning of certain technical terms. Any copyrighted or trademarked terms or abbreviations are used for educational purposes and remain the sole property of their respective owners.
brought to you by http://geekofthehouse.blogspot.com