Security with A.I.
Options
Comments
-
Thanks for taking the time to type all of that out, John. Honestly, I am more worried about someone stealing my identity and taking out 5 mortgages on my house, then finding out near foreclosure that it happened.
Most of my audio purchases these days are made by speaking directly to the designer or manufacturer, and I just buy direct through them, sometimes at a reduced rate. So I don't see how AI could infiltrate an already agreed upon price. Most all of my groceries are through Sam's Club (fixed pricing for business members), so while they may get me on some items, the most expensive ones (audio purchases), I don't see how they could affect pricing, although I may be nieve to think that. I just don't know how they could alter the prices.
I have no smart appliances. But the guest coming over aspect was enlightening. That is definitely a vulnerability!
Question - would it matter if they were only allowed on my guest network? Or are we just as vulnerable?
Tom
OK, so, this dynamic pricing is not everywhere...yet. The vast majority of stores are not doing it...yet.
For it to work, you need to go through a retail front. A direct buy from a manufacturer or something is not likely to be affected. Mainly because they don't have the resources to implement it.
Dynamic pricing already happens, though. It's also referred to as "surge pricing" and you see it happening the most obviously with air fares and Uber pricing. High demand surge pricing has been around a long time.
But the newest iteration is called algorithmic pricing and it is a form of dynamic pricing that follows "algorithms" that "follow" you. With facial recognition and digital device promiscuity, they can dynamically identify your personal profile, establish your base algorithm and change pricing based on that. So if they think you're rich, they'll bilk you out of a few more dollars.
The biggest concern isn't so much the pricing shenanigans but rather, how the hell did the Kroger get your personal data to know how much to shaft you on that 3 pound block of Velveeta you got your eye on?
There's been legislation put in place about it already:In November, 2025, New York state's "Algorithmic Pricing Disclosure Act" came into effect. The law requires companies to disclose when prices are set using a consumer's personal data. The bill enables civil penalties of up to $1,000 per violation.
So basically, NY requires stores that use this type of pricing to disclose that it is in effect.In April 2026, Maryland Governor Wes Moore signed HB0895 into law, which bans grocery stores and third-party grocery delivery services from using consumer personal data to raise prices for consumers. The law authorizes $10,000 penalties for first-time violations and $25,000 for repeat offenses. The law is scheduled to take effect in October 2026
Maryland basically outlawed it completely. Several other states are following this tact.
The personal algorithm pricing is still a ways away from being really prime time. It's mostly been in a test and see phase with targeted markets with a diverse population like NYC.Retailers, and online retailers, in particular, adjust the price of their products according to competitors, time, traffic, conversion rates, and sales goals.
You already see it happen at Amazon which isn't so flagrant but if you shop TEMU or Alibaba it is flagrant. A good example is the Best Buy website "anomaly" where the price is higher in-store and when you tell the store personnel it was cheaper online, they look it up in-store and they go to specific web server that shows the in-store price, not the non-Best Buy store site which shows it cheaper. We used to call this the bait and switch.
It can be beneficial, though, as places with perishable products can adjust prices to move them faster so they don't have waste. A spoiled product that didn't sell is a 100% loss. A product that isn't spoiled but sells for a 20% discount is only a fraction of that loss.Supermarkets often use dynamic pricing strategies to manage perishable inventory, such as fresh produce and meat products, that have a limited shelf life. By adjusting prices based on factors like expiration dates and current inventory levels, retailers can minimize waste and maximize revenue. Additionally, the widespread adoption of electronic shelf labels in grocery stores has made it easier to implement dynamic pricing strategies in real-time, enabling retailers to respond quickly to changing market conditions and consumer preferences.[35] These labels also make it easier for grocery stores to mark up high demand items (e.g. making it more expensive to purchase ice in warmer weather)
So there's nuances to it and it's been around a while. The new AI and massive processing data centers are making it easier to get things into real-time but, it's not there yet. But it has the potential to be very damaging even if it isn't abused so it's good to see legislation establishing boundaries for it already.
As far as the network stuff, a guest network is typically separated from the main network and there's no way to jump between them. So the entire idea of a guest network is to secure your home network from visitors who might bring threats into your environment.
Expert Moron Extraordinaire
You're just jealous 'cause the voices don't talk to you! -
Thanks for taking the time to type all of that out, John. Honestly, I am more worried about someone stealing my identity and taking out 5 mortgages on my house, then finding out near foreclosure that it happened.
Freeze all your credit. After the Equifax breach in 2014 (IIR), we froze all of our credit. even with lesser known bureaus.
Equifax, Transunion, Experian, Chex systems, and Innovis. There may be more I forgot.
I can't even get my own credit report without unlocking one of them.
That's enough to stop anyone trying to get a mortgages or car loan in our names. Not much else you can do.
That's a bit excessive.
Sure, there's lots of people out there like Dave Ramsay who swear up and down that credit cards and the credit market serve no good purpose for the average joe and they should exit post haste live by the cash is king mantra. But, unless you are independently wealthy and can pay cash for everything, you need credit.
Freezing your credit does more damage that one thinks too. Nobody can open a new account under you but nobody can report on your old/existing accounts either. No big deal until you need to unlock your credit. Sure, it means you have your credit available to you again but it can take as much as 6 months for any recorded credit history with your current creditors to be available to new creditors on your credit report. That ends up being a big delay or even outright denial if you need to get a line of credit from something as simple as a store credit card to a mortgage, it can really kill the deal on you. For example, storm damage destroys your roof, insurance company denies the claim says you didn't have a "wind damage rider" so you need $20K like yesterday. Hey, bank of cool guy, can I get a loan for $20K. Sure, dude, lets take a look at your cre....welp, it's frozen, that's a no go. Sorry, man, unfreeze your credit and we can help you. UGH! That's going to take MONTHS and my roof is like swiss cheese NOW! Oh well, too bad, so sad, sucks to be you, exit is on the left, honcho, have a good day.
Credit freezes weren't meant to be a permanent way to handle credit. They often come with flags on your credit report so that if any attempt, even you, happens where a line of credit is attempting to be established, it'll start a legal process and potential fraud investigation. Credit freezes are intended to be used when you believe that there is a real and present threat to your credit. Something like a lost wallet internationally, a potential stolen identity or some level of fraud like a spouse or business partner trying to establish credit in your name.
Freezes, though, don't stop fraudulent mortgages from being issued on your property. That's deed theft and it's not as prevalent as certain commercials would have you believe. What deed theft is concerns the named owner of the house, not their credit. They forge some documents or get you to sign fraudulent documents under duress and then go file a deed transfer, often times a "quit deed" which requires no financial information at all, it's pretty much claiming a property in abandonment and exists so banks can get out from under a foreclosed property. Anyway, they get your deed registered in their name and then take out mortgages on it, take the money and leave you hanging. Your credit doesn't get touched, at all. Most people don't know it happens because the nefarious evil doer has the mortgage docs sent to a different address and you don't even see it on your credit report. You know when the eviction notices and foreclosure notices start coming to the house because they are required by law to send that stuff to the property location address as well as the business address. So a credit watch would never show a deed theft. Since it's near impossible to do on a mortgaged house underwritten by a bank, you wouldn't even get a default notice on a mortgage because the mortgage was never in your name just against your property as collateral. Is it on the rise? Yes. But, in 2023 there was a little over 9500 cases nationwide. You hear the 600% increase number but that's from 2015 to 2026. It's only been on the rise because consumer level computing, optics and imaging tools have gotten significantly better. Between 2019 and 2023 there was less than 60K cases nationwide. That's not even 12K a year. There's 147 million houses in the United States. So yes, the number has increased but it still substantially rare in the grand scheme of things. Of those title fraud cases, less than 10% were done against an occupied house and they involved some level of phishing and deception on someone like an elderly owner. The vast majority were on vacant properties, properties in probate due to will disputes or other inheritor negligence on dealing with the property, foreclosures and rental properties that are seasonal and vacant for a significant portion of the year.
Additionally, States have procedures in place that make deed theft difficult. Some States, like New York and Michigan, have more lax requirements for things like "quit deeds" that make it easier to forge and transfer the title than others but many states require, also, a tax transfer form that requires direct, in-person, notary witnessing for both parties of the deed transfer. Additionally, the feds and every state has services for tax notices that you can sign up for for free. Does this mean we're all safe? Nope. You should take every precaution you feel is necessary but if you have a mortgage, you don't have the deed to your house so you don't need deed/title insurance. In fact, it's probably part of your PMI, read your fine print. The bank does have your deed and they are on top of it. If you own your house outright, no mortgages or liens, get title insurance.
As far as what you can do without a freeze, it's simple really. Every credit reporting agency has notification capabilities now. Sign up for it. You will get immediate notification of any credit report request. Check out your bank, see if they have credit monitoring services for you. 'Cause it's way cheaper for them to pay for the monitoring service for you than to eat the fraud that was foisted upon you.Expert Moron Extraordinaire
You're just jealous 'cause the voices don't talk to you! -
This guy actually proves the pricing variance. He calls it surveillance pricing.
https://youtu.be/tCSwx9LZtD8?is=3xQLK_3Sbo5Fq0Pb
Finally had a chance to watch this video and it was actually a really good video. Guy demonstrates how the surveillance pricing works and I have seen this first hand myself.
We got the Navelgazer in 2023 because of this baloney too.
So the last time we flew to see the fam in Midland, TX, it cost about $4500 round trip for 4 people for just flights. Ouch. Additionally, a rental car while we are there cost about $3500-$4000 'cause it's a "low capacity market". Not many people vacation in West Texas oil country, I guess. So, anyway, $7,000-$8,000 a trip 'cause we stay for 2-3 weeks.
Navelgazer costs about $600 in gas round trip and we spend about $750 in total on food and hotel rooms for the 4 of us on the 24-26 hour drive. So a fraction of flying, significantly. Figured if we got a used vehicle that hauls everything we needed for about a month in Texas, we'd save a bunch. Since 2023, we've done 5 trips to Texas. At about $1400 a pop, we've spent around $7500 on trips since June of 2023. So 5 trips for the same price as 1 flight and rental car. If we make one more trip to Texas, the Navelgazer's sub-50K price will have paid for itself in savings from not having to fork over the ENTIRE yearly vacation budget for one trip.
Anyway. I told you that to tell you this. While I was doing all that figuring out, I noticed something. Jackie will usually book a hotel while we are driving so we can adapt to weather, traffic, etc. on the road. I will choose a stopping point and look at prices for hotels in that stopping point and if they are more than we want to spend, I'll check another stopping point for better rates.
So using Lebanon, TN as an example, I will look up rates there and the cheapest hotel I was able to find last trip was $237 a night for a double queen bed room. Jackie, on the same app, same time frame, same room paid $207 for the room. The only difference is who looked up the price. I have a much more extensive traveling history than Jackie because of work. So my "profile" means I'm "willing to pay more". I'm not, actually but, what stinks about this is that I might be willing to pay more for a better hotel room but I don't have a choice to not pay more for any hotel room.
I also checked flights too. Jackie and I, both on the same network, in the same location, at the same time, looking at the same booking site for the same flight, same days, times and everything the same. Jackie's round trip price, $5300. My round trip price $6900. That's a 30% increase for me.
This has been noticed since we started trying the drive to Texas in 2019 because the plane tickets in 2018 were $3600 round trip and that was getting unpalatable. We also rented an SUV while there and that was $1800 for the week and a half we stayed.
The difference between the surveillance pricing that this guy is talking about and the new algorithmic pricing is that they can only do this surveillance pricing online, easily. Like the Target example in the video. You shop online and do delivery or curbside pickup you are stuck with the surveillance pricing they foisted upon you. However, you, yourself, walk in to the store, get your stuff and pay in store, you're not getting surveillance pricing.
What big retail is trying to do now, though, is to use those new carts where they scan and price stuff as you put it in the cart and then they charge your account? Yeah, you need an app to use that and the fine print of that app says they are mining your personal data. So they will do that surveillance pricing in the store too. Which doesn't jive with most consumer protection laws where a price is supposed to be clearly marked to protect consumers from being fleeced like this.
I have seen systems demo'ed, though, where they use facial recognition to get your "profile" and things ring up at different prices for different profiles.
So it's easy to change the presented prices online because they make the change before the product page loads so you only see your tailored price. But in person, it's more difficult and they have, apparently, found ways around that.
The massive amount of compute power behind A.I. is what is able to drive this too and the fact that A.I. can do this profile work exponentially faster than humans. But without your data, it doesn't work.
That's why they want you super social on social media and arguing about anything and everything because you give away a lot of what makes you you when you get into heated arguments with bots that are designed to just tick you off and get you going.Expert Moron Extraordinaire
You're just jealous 'cause the voices don't talk to you! -
I refuse to use these so called store apps to get "digital" coupons to save .30 cents, Fvck it, I'll pay the .30 cents. The amount of intrusive crud that comes along with these so called "apps" is over the top. I do not need more stuff in the background mining information about my personal purchases or buying habits. Heck I've had to stop Sams Club from sending me "It's time to buy these things now" I started getting stuff sent to my email account telling me I need to buy more or that they have already compiled a shopping list for me. I do not need your help making a list of things I need, NOT things you think I need. Besides I go there to get items in bulk that I do not need to worry about gong bad. They seem to think I need the conga huge jug of laundry detergent every week. The damn thing lasts me more than a month.What big retail is trying to do now, though, is to use those new carts where they scan and price stuff as you put it in the cart and then they charge your account? Yeah, you need an app to use that and the fine print of that app says they are mining your personal data. -
So, saw this this morning, reported on yesterday about a Palantir sourced and managed command and decision AI.
https://news.slashdot.org/story/26/09/22/0614238/pentagon-investigators-say-overreliance-on-palantir-ai-contributed-to-us-strike-that-killed-123-iranian-children
There's a link in the article to a Bloomberg investigation and piece from The Los Angeles Times too but both are probably behind pay walls so if I share a link, it'll mess up any free access you might get. So go to the Slashdot article and click on them yourself.
That AI is called Maven or "Project Maven" and it is an artificial intelligence-driven command-and-decision platform used by the U.S. military to process battlefield data, coordinate intelligence, and accelerate targeting decisions. It's provided by Palantir. It's basically used to collect target data from multiple sources and combine and collate the data into a targeting and engagement plan faster than humans can so time to decision and time to target is reduced.
How it ended up making a mistake is not really going to be known for obvious reasons. But this happens when humans rely too heavily on AI without checking before acting. If anything it was human error in putting too much trust in the AI and not checking the intelligence before engagement.
This is an extreme case but many decisions get made putting far too much trust in AI systems without any oversight or quality checks. It's becoming a big problem in "vibe coding" where people are using AI agents to do the actual coding and the human "coder" lacks the necessary skills to review the code and perform proper regression testing before publishing it. It's getting so bad that places like Github and Source Forge are putting together teams of volunteers to peer review the "vibe coding" before release to try and head off any egregious errors.
These biggest problem, too, isn't the failures from poorly constructed output or a lack of capable oversight but rather that much of this "vibe coded" stuff is full of security problems and many create exploits that other AI systems or nefarious evil doers can exploit. That, right there, is what is at the forefront of AI. Not the rapid expansion of the AI intelligences and such but the security around them. It is currently a rapidly emerging market that is already booming and it's not getting smaller any time soon.
So if you have an up and coming college freshman who is looking for a field to get in to, data security and AI security are going to be bigger than they already are and companies are paying top dollar for people who are good at it. So if you got a young someone who is saying they want to "get in to I.T." then tell them to start figuring out how AI works and start learning about the security concepts and tools to manage AI. Then go get a degree in network/data/AI security and they'll be making so much money right out of the gate, especially if they are capable at it, that college debt won't last but a couple years.Expert Moron Extraordinaire
You're just jealous 'cause the voices don't talk to you! -
https://share.google/DNfZQmleedhgMLfry
I found this interesting and relatable.
Here’s a summary if you don’t want to click on the link.
Things work out best for those who make the best of the way things work out.-John Wooden -
YEP WE'RE DOOMED !!!
-
AI is going to be renamed “Super Intelligence” in any official document.
For real.Since we don’t know where we are going we have to stick together in case someone gets there….Ken Kesey.
Idolizing a politician is like believing the stripper REALLY likes you.
And the boots you lick will stomp you too. -
Heck, I can talk about grills 3 times, with my phone turned off, and the next time I log on to FB? I have 4 ads for grills on my feed.
I apologize for not reading the full thread yet, but I did want to chime in on this. Everyone assumes it's YOUR phone listening to YOU all the time. First off, you'd be surprised how much information is available without listening to you. The importance of proximity and network tracking is huge here.
So let's take your example. You talked about grills 3 times. I'm assuming you had the conversation with at least one other person. Possibly up to 3. Maybe that other person has multiple devices tracking their audio. Maybe they were looking at grills just before, or just after. Maybe they were looking at grills while connected to your home or office wifi. Are you Facebook/Instagram/Forum friends with this person? And so what if your phone was off. What about your computers. Or TVs. Or a range of other connected devices that you have to sign into to use. All of those things have profiles. All track IP addresses. All can elegantly stitch together who you are and who you've been with. The good news is, none of them care about you, Tom the person. They all care about customer xyz123 has taken actions abc which places them in a higher intent for potential buyer profile. So what are actions abc?
Guess what, the logs show that you two were within proximity of each other. Even if your phone was truly off, if you shared a network the IP can show that. If you didn't share a network, it can still show these people were in proximity with each other with a range of other devices. Friends on social media with any recent interactions? Ya'll must have talked about any of the things you've been interested in. Any of these things can trigger programmatic advertisements and content to essentially say "if his friends in close proximity of this is currently interested in grills, he might also be currently interested in grills." So the software and the immense amount of data available will put all the pieces together, drop you into a targeting list, and now BOOM, you've got some grill ads in your feeds. Didn't matter if your phone was off.
You really want to prevent this? You need to have all your conversations deep in the woods, fully off grid, no devices allowed within 10 miles and no internet use of any kind for 10 days before or after the meeting in the woods. Good luck with that.
Analog: Hana ML > VPI Prime Signature 21 > Bob's Devices Sky 20-S > Sutherland 20/20 w/LPS
Digital: Cambridge CXC / Streaming > Cambridge CXN v2
Norma IPA-140 > Rosso Fiorentino Volterra
Hifiman EF600 > Hifiman Arya Stealth
Discogs -
Eh, targeted ads are really tied to you. Yeah proximity helps in deciding that but the correlations you are talking about in regards to other people around you and their devices is an egregious violation of privacy laws.
The proximity stuff that does happen is things like you logged in to a WiFi at Lowe's and your data tracking now knows you are at a Lowe's.
You don't even have to be talking about or even looking at BBQ grills in Lowe's either. You could have just gone for a furnace filter and some screws, totally unrelated to BBQ grills. But if it's the end of August and you watched a TikTok video of a guy making something on a BBQ grill the day before so, you have a statistical probability of better than 50/50 to be taking advantage of an end of season sale to get a good deal on a new BBQ grill. Targeted advertising from Lowe's or a competitor has a better than zero chance of scoring a sale from you. That's how the proximity stuff gets you.
Your buddy's phone might affect him if he's not looking for a grill but you were talking to him and having a conversation about the killer deal on a grill you just got. If he says nothing, he probably won't get anything targeted. If he says anything, especially in the positive or showing interest, he's getting targeted ads.
Hell, if you're reading this thread on your phone right now, you're going to get targeted ads about a new BBQ grill.
But the location tracking and the browsing history has way more to do with it than your conversations.
That said, everything with a microphone is most definitely listening to you. Or at least someone. Your TV, unless you have voice identification, is going to be fairly indiscriminate. Your phone, though, that's mostly going to be you and nobody else. It can tell the difference between you and anyone else too. When stuff like Apple and Google first started doing the voice commands, you could listen for someone saying "Hey Siri" or "OK Google" and then run up and yell something like "HORSE ****" and affect their "algorithm". Realizing the security risk that posed and how it made them liable for privacy violations, now Siri and Goo...I mean Gemini require you to set up voice recognition to you and only you so you can't be compromised by some rando on the street.
Doesn't always work, though, 'cause a Google commercial or a T-Mobile commercial will make my stupid phone start listening for something. Then I'll get ads to upgrade my phone.
So your behavior and how you use your devices does far more to affect your targeted advertising than any surveillance features.
For example. My phone knows that sometimes I make it listen to ambient music because I'm liking what I hear and I want to know what the song is. Whether I engage the listening app or not, if I go to YouTube while I'm in earshot of the music, YouTube brings up the artist and song currently playing as a Suggested For You option. If I actually use the app to listen, I will get inundated with ads about that artist, that artist's social media pages as suggested for you, any media appearances, tour dates, ticket prices and album releases on all kinds of media from YouTube to Facebook to Google results ads.
The bad part about all this is that this aggregated data is available for data brokers to buy from Google or Apple. Those data brokers are REALLY good at making statistical predictions from that data because when they can give good information based on that data, they can charge more for it. It's really hard to keep on top of that too 'cause every time you do something, anything that looks into your background, that data gets puked up about you. Once that data gets puked up, whomever was standing there with the bucket now owns that data about you. They are obligated by law to protect it and only hold on to it as long as they need to but while they have it, they are allowed to not only target you with their own ads but in many cases, they can share it with 3rd party partners to also target you with their ads...or phone calls...or text messages....or spam, both sneakernet and email. Sometimes, those 3rd parties are less than scrupulous about keeping up with data security and not reselling it as their own packages. Sometimes they get breached too and than that data goes to the nefarious evil doers and all bets are off on who gets it, when, where and what they do with it.
If you want to see how this trickles out, go apply for a new credit card. Or better yet, just sign up for a brokerage account at, like, E*TRADE or Schwab or something and then marvel at the numbers of BS outreaches you get from everyone and their brother.Expert Moron Extraordinaire
You're just jealous 'cause the voices don't talk to you! -
@JstasBut the location tracking and the browsing history has way more to do with it than your conversations.
Question, If your phone location history is turned off (which mine always is) and you use duck duck go and set it to remove history as you close how does this affect things? Recently in the last month I've noticed that if someone sends me a link or there is a link say here for yaba yaba do do on youtube, when I click the link to view I get a pop up from YT stating I have my history turned off and it will not load said video or music. It will ask my to enable history to view, which I decline and go about my way.... -
Location history is not the same as location tracking. They will still know where you have been from cell tower pings and other RF scanners that pick up passing Bluetooth signals. The stuff is so complicated, they can even identify different passengers in a car that drives by.See profile for rigs. Nothing else here.
-
Yeah, what Bill said. That's how they know if you're at a Lowe's or not. What's really wild is that Lowe's and Home Depot both provide "in store maps" on the website and they are geolocated so they can actually tell what aisle you are in from your location data. But that level of location tracking is not available everywhere...yet.
But for your phone to be listening to someone else's conversation and identifying that someone else and then targeting them because your phone surveilled some guy you don't know from spit, that's a location tracking and proximity thing that is not legal. Does it happen? Probably. Is it legal? No. Clear violation for surveillance and eavesdropping. Even if it is in a public place, unless the owner of the device that did the eavesdropping gave clear permission to allow the use of that device for that kind eavesdropping surveillance, it's violating the rights of both the device owner and the person being eavesdropped on.
But also, I use duckduckgo as well and I've turned off much of my ad stuff so I get the most random ads on stuff like YouTube or Facebook.
I know they require location because they have intellectual property laws and some stuff on YouTube isn't available to be viewed in your location, etc etc etc
As far as the history stuff preventing you from viewing, there's no real reason for it to be a control criteria beyond tracking trends so they can sell ads. I don't know exactly how YouTube's ad algorithm works in the nitty gritty details so there might something more there but unless your blocking more than just history like location and age, there's no reason for viewing history to stop a video from playing outside of forcing ad revenue.
But YouTube knows when you have an ad blocker and they make it difficult to use YouTube when you have one. Which is funny 'cause Alphabet Inc owns YouTube. But they also own Google. And they own Chrome. Chrome has ad blockers built in. So they stop you from using their product when you are using their products to access the product they are blocking because of a feature of their own product.Expert Moron Extraordinaire
You're just jealous 'cause the voices don't talk to you!




