It likely doesn’t take too much for any group of coordinated individuals to significantly sway any social media. I mean, even on lemmy – if you had like 5 people who scoured ‘new’, or who had a program to scour ‘new’ posts for specific keywords, and just downvoted/upvoted specific content, it’d have an impact. Enough downvotes will hide posts from most other users, enough upvotes will amplify the message.

Similarly, if you look at who posts to lemmy, there are some folks with like 10’s of thousands of posts in a year, and those posts often follow specific ideological agendas. I once asked one of these sorts of accounts what their situation was, wondering if they were like a paid employee of a PR firm or something given that they had over 50 posts per day, and the response was “You have no proof”. I mean, a simple “Nah, I’m just passionate about certain topics” would’ve sufficed, but instead they chose to deflect the question and claim innocence via a lack of definitive proof. If I had ‘proof’ I wouldn’t have been politely asking/curious as to their response.

Social media is basically designed to be manipulated by propagandists. Doesn’t really matter what corporate structure is behind it, whether its federated or not.

Do you think banks and credit card companies / credit bureaus aren’t already using LLMs on their massive amounts of data?

I just recently had an issue with Access, where running a report multiple times would randomly result in errors, or clean reports. The explanation copilot gave was that Access has a built in LLM now that is “interpreting” certain queries, and may result in different outputs, even different resulting data types. So doing a join on those outputs for the report, would sometimes result in reporting errors, even without changing the underlying logic. AI has sorta taken a calculator that used to reliably return 2 + 2 = 4, and turned it into one that sometimes thinks 2 + 2 = chair.

I don’t know how comfy I am seeing AI integrated everywhere at this stage of the tech. It really doesn’t feel mature enough for production.

Ah, makes some sense. The mobile networks are even more erratic with how they assign IPs – though I’d still be a little surprised if it was the wrong country entirely. It’d imply the provider is using IPs from a single range in multiple legal jurisdictions, which’d inherently make things like geofencing more difficult. Sorta like VPN functionality to access foreign data regions, as a result of sloppy configuration and negligence by the ISP. Wonder if it could also be something to do with IPV6 – I think that’s more common to see amongst mobile networks, and I’m honestly not too sure how well that can get mapped to geo locations – I’d doubt the site, how its put together, would be tryin too hard to sort that out.

While I don’t disagree that most countries ought to be looking for what you imply, I think there’s a somewhat interesting counter point to it. The anonymous aspect of online interactions/actions has contributed to some really negative trends.

One easy recent example: private citizens in the Netherlands, having taken a course on ‘faceless influencing’, used youtube to push out a bunch of videos encouraging the breaking apart of Canada, while pretending to be Canadians. They allegedly did it because the rage-bait clips generated lots of clicks, and ad revenue paid to them by Google, a company which also worked to amplify their fraudulent personas – a company that’s also tied closely with America, given the administration’s ‘tight’ ties with their tech oligarchs, a country which has overtly expressed interest in breaking apart Canada. The anonymity of those users basically allowed America to crowdsource their nation-destabilizing work.

With AI junk, that’s all going in to over-drive. The ability to fake ‘people’ is likely a big part of why govts have changed their tune on trying to ID people online. Demanding transparency is likely viewed as almost a necessary evil to combat the deluge of propaganda that’s coming out of places like America these days.

There’s also the general tone of online interactions, and the breakdown of certain social-communal parts of offline-life. “The male loneliness epidemic” for example, being partly born out of guys being captured by the manosphere, or being isolated ‘in real life’ due to their excessive presence online. Also the generally combative tone most people take on sites like lemmy / reddit – which pairs with the structure of most media content, which tries for sound bite click bait rather than nuanced constructive thought: you get way more upvotes, if you respond to a small out-of-context shock-value clip with your own short one liner type rebuttal, than if you actually engage with the other parties comment more genuinely.

Like the people generally demand transparency on government actions/decisions, because having those decisions exposed, and the people making them accountable for their actions, helps to reduce corruption / bad behaviour. Same general principle when anonyuser204956 is busy spouting nazi shit, if suddenly their friends/family/coworkers can easily see what they’re doing. People keep other people in check.

Hm, wonder why that’d be – it implies heavily that it bases the country on the IP address, which in theory is done by looking at what company the address is registered to, for the most part. Like I’m guessing it got my city wrong, because it used an address that the ISP provides for the IP range, which isn’t the same as the city I’m in, because the ISP uses it to cover numerous cities around the broader region. I reckon if you’re using something like Starlink, or other similar international-ish provider that may be very loose in how they associated addresses, it’d fail most times.

Heheh, a whole lot of mocking in this thread, but I don’t mind the site / its display.

Yeah, it’s overly melodramatic in its setup, and a bunch of the information doomerism is silly in terms of the info basically being required to provide data comms etc. It also tends to get things a bit wrong in a few categories – like for me, it said I was in a totally different city (still the right country at least - Canada), then it said my time zone was in iceland, which is kinda… no.

But the general message of the site, and the awareness its trying to raise in regards to how much data gets shared for basic comms establishment, and how that information gets used to fingerprint people, is worthwhile.

So… they’re taking advantage of one of the selling-features of streaming services, that being the ability to scale your spend pending your own personal preferences. In the early days of streaming services, they had enough content to justify paying a monthly subscription each month – it’s not the customer’s fault that streaming got enshitified. Hell, a bunch of them switching to ‘weekly episodes’ was just a very poorly disguised attempt to drag out how many months they thought their one flagship show could capture audiences. The old practice of dumping a whole season all at once in one month, because you knew you’d have another season of some other good show the next month, is practically gone – with streaming reverting back to the old network practices they’d usurped.

Same with games. Tons of titles are just shitty early access things, things that get abandoned mid-development, things that rely on a live-service platform that companies’ll shut down a month or two later, and so on. And some titles are askin like $80+ for their shitty offerings. Yeah, that’s not the customers fault in the slightest. They’re right to look for discounted offers, what sane person wouldn’t?

Yeah – I think a lot of people who took even just one stats course are in a similar boat. Though I think it’s a bit easier to understand the shift if you frame it within the context of Social Media sites controlling the population’s opinions / propaganda.

Most govts understand at this point, internally at least, that if a message is repeated often, loudly, and it saturates a people’s media, they start to believe it / agree with it. The survey, and the reporting storm surrounding a survey, isn’t so much about showing people an accurate representation of how people’s viewpoints vary, but rather a vehicle for govts/companies to tell people how to think. Sites like Facebook don’t so much as sell advertising, as they sell the ability to socially engineer its users to like your product / political stance: make enough general noise about a niche position, and people will think it’s a majority opinion.

Where the bots get used in the workflow, isn’t really that big of a concern.

Seeing as the vast majority of “polls” I’ve seen in the last 5-6 years have been “this was a poll done online, so we can’t assign any certainty or margin of error, cause we have no idea who actually responded and it could’ve been just like, two dickheads with bots spamming nonsense, but the results were click baity enough for us to run a story” … I don’t see how them cutting out the two dickhead middle-men, and just using their own bots, is really that much different.

A potentially odd thought, but the power and reach of US tech oligarchs is based on US softpower – the same stuff this guy is basically saying is useless. One reason US stocks/companies get lots of investment/support is the belief that US companies have more default in-roads to all western nations – invest in a US tech co, you get access to all western markets. It’s one reason they have the market power they flex. Even Iran had reports of their cisco devices failing during recent US aggression – so even the USA’s enemies had figured US tech was ‘country neutral’, stupidly, and their obvious mistake cost them dearly. Other nation’s will see that as a learning moment.

Western nations have started pulling away from US tech / hegemony. Places like France are eye’ing Linux, the Netherlands central bank declared it reasonable to expect US tech excised from their banking ecosystem in 4 years or so, Canada’s openly declaring US ties a weakness. So investing in a company like Microsoft, may no longer translate into investing in a company with a global reach within the western world – their stock ‘should’ eventually price this in. It’s one reason the tech overlords have directed the US government to challenge any/all pushes for data sovereignty.

And as for the end of nuke dominance and rise of AI drone warfare and all that… he’s an AI hammer salesman, declaring all problems are AI nails… hes clearly biased. We’re seeing drones as a viable option currently, because people are so scared shitless of nuclear conflicts. Just because people are too scared to use the massively destructive weapons, doesn’t mean that the deterrent factor of those weapons is meaningless. It’s why few countries have directly helped ukraine in their conflict with Russia. Hell, the states and Israel are busy justifying a war to try and prevent a country from getting nukes – and they’re beating up a country that only had drone power. While their campaign isn’t going as well as they may claim publicly, I know Iranians who’ve confirmed things like “All the airports are destroyed, and most of our projects/work is closed, so we’re just sitting at home kind of waiting at this point”. Those drones didn’t really help deter anything / protect the people all that much – having nukes likely would’ve, as we see in cases like North Korea and Russia. Even in cases like the USA, where their veering into a fascist dictatorship garnered little comment from western nations, who were afraid of upsetting their nuclear-umbrella.

I imagine a system where the government just prints cash, would have a whole slew of other problems and issues, and would likely be less stable/viable than the current mess. The current approach has flaws, but if managed properly, remains viable for a long time – most of the flaws that become existential threats, only show up after years of neglect or willful corruption, such as in the USA.

Building on US tech means the US generally has control over whether you can deploy your military assets, and gives a foreign, militaristic/fascist trending power, deep insights into your military operations. Pretending like these risks are not greater than, or at the very least on par with, “its hard to integrate systems/build our own” is silly.

It’s sorta like Canada’s former liberal leadership hopeful Chrystia Freeland acting like China’s the biggest threat to Canada. While the US administration is actively and openly trying to dismantle Canada using economic warfare, is ignoring former international conventions like those pesky ‘human rights’, and so on. Like there’s this old joke about Canada being in bed with a sleeping elephant given the disproportionate sizes and risk of that elephant rolling over and accidentally squishing you. Except the elephant woke up now, and is actively trying to harm you. Meanwhile idiots like Freeland go on about some Chinese Bear that’s a threat primarily on the other side of the world, ignoring the elephant in the room.

The USA is a threat. They are actively attacking anyone they feel like. They are actively antagonistic towards their “allies” and neutral nations. Their tech oligarchs actively talk about setting up their own baronies, aka “Freedom cities” in the hollowed out carcasses of what remains of nations. Their state department actively opposes foreign nations pursuing data sovereignty, because the USA doesn’t care about privacy, especially not for non-republicans – they want that data to target “terrorists” (non-republicans) more easily with the use of AI. Their leadership quite literally called all their Generals in to a room last year, said “We expect you to commit war crimes, cause we want the world to fear you” and fired anyone that objected. The USA isn’t just a ‘risk’ of being a threat, they are an active threat undermining western democratic nations. Why anyone would think there’s a greater risk ‘not’ to give these folks more power/control over you, is beyond my understanding. My closest approximate comparison on a day-to-day relatable level would likely be something like an abusive relationship, where the victim rationalizes staying in the relationship because “If I left, they’d outright kill me”. That ain’t healthy, nor a desirable position for a military.

Like even the Iran / Hormuz stuff, is basically intentional pain inflicted on the EU. Last year, as part of their chat leak during their strikes on yemen – the chat that leaked on whatsapp or whatever – Hegseth, Gabbard and them were complaining about how they felt they were policing the area, even though all the benefit went to Europe in the form of open trade routes. They wanted Europe to be more actively involved. Trumps made clear references of a similar nature, with his regular bravado/crassness, in his recent “we probably shouldn’t even be there” comments.

The current US administration also has a focus on isolating opponents (which they tend to talk about as ‘containment’ in their ideological writings if I remember right). It’s what underpins things like what they’re doing to democrats in places like Minnesota, and building concentration camps for “illegals” (non-republicans, and non-whites) – they want enemies isolated, cut off from outside aid. Even more, they want those people to suffer, and make noise as they suffer, as it helps to keep other blue states in line and lets them point at the suffering to appease their base. A similar approach underpins much of their international relations, cutting off nations from trade opportunities to weaken “opponents” (non-nuclear / smaller nations) – see Cuba as an easy example currently, or the ongoing attack on international trade norms. Attacking Iran cuts the EUs oil supply (among other trade gaps), exposing a strategic weakness and providing greater opportunities for the US to sow discord amongst EU block members: enter the relaxing of Russian sanctions to further sow animosity, as some EU nations are pressured to resume Russian trade. Trying to distract from his Epstein atrocities is part of the reason Trump may’ve agreed to the plan and rushed the timing a bit, but pretending like it’s the only reason for the current shit going on is naive – there’s a whole fascist administration, full of out and proud Christian white nationalists, backing the actions of Trump, and using his antics to distract from their goals.

Rubio literally sent out a memo in december if I remember right saying to aggressively counter any tech sovereignty pushes, as the trump admin wants access to all foreigner data for AI integration and “national security” of the USA. They want to hold/have access to it, cause they like using it as part of their AI surveillance and snooping regime. Again, if I remember right, that was circulated to embassies and lobby firms etc etc.

So any news story about how hard it is, is likely a US influence campaign. Using their oligarch control of media to magnify issues, think tanks publishing unprovoked ‘white papers’ that support the US narrative, and on and on.

Yeah, but this breach is specifically about KYC, about financial industry stuff. The company that got porked, was the company the banks used for their KYC stuff.

KYC is typically a due diligence process tied to regulated financial industry participants – the restaurant example has a much different function. Banks and FIs have much broader retention (and disclosure) obligations.

Here, let’s put it slightly differently. I’ll reference Canadian regulations/processes more, as those are the ones I’m most familiar with. If you’re a bank, you’re required to flag suspicious transactions related to the customer – and in order to know when those transactions are suspicious, you need some way of reviewing it within the context of the customer. You may even have an obligation to second guess / question / try and advise the customer ‘not’ to make a transaction, based on knowing your customer.

The most basic example of that, is where Credit Cards will decline payments / request a call if you try and make a purchase in a totally abnormal location – like you “know your customer” lives in Toronto, but suddenly see them spending money in Mexico? Or if they called you before they took a trip to mexico, that’d also go into a KYC type file to let people know to expect those sorts of charges and let em get processed. That’s tied to KYC.

The media will often run stories about seniors getting scammed, with the general message being “WHY DIDNT BANKS DO MORE TO PROTECT?”. Well, that’s KYC too. You gotta ‘know’ your senior members, and their spending habits to some extent, to find those outliers. You also need to be familiar with them enough to know whether its “normal” for them to come by and take out cash, and in what quantities and for what purpose, cause seniors will sometimes ‘show up’ with a person pressuring them to take out cash to ‘pay a bill’ (scammms galore!). All part of KYC due diligence.

Or the somewhat obvious elephant in the room – if you have a “personal” account member, who keeps receiving etransfers to his “[email protected]” account for some reason, you gotta look into it a bit and sort out what all those payments are related to, cause it isn’t a business account. And if you see anything suspicious, it gets reported to the authorities, where, most likely, Trump shits himself and Americans ignore the crimes.

People want surveillance options. One of the highest/most obvious features required, unsurprisingly, is the ability to see your cameras on your smart phone – which generally means you need a Smartphone App + a centralised server/system connecting the different ends. The alternative being that end users would likely need static IP addresses / Dynamic DNS setups to have a Smartphone app point “directly” to their exposed CCTV ports – which I don’t imagine regular consumers are keen on, likely why basically no such options seem to exist in the retail space (afaik - if there are widely used brands i dont know about, by all means clue me in).

Options that are fully local/closed/under user control, are almost impossible to find. This isn’t so much a consumer-specific problem, from my perspective, at this point – there aren’t enough options for consumers to choose differently. It’s sorta like how you’re generally ‘stuck’ with US-tethered Smartphones. It’s not so much a ‘choice’ that consumers get to make, as it is that these big businesses have effective monopolies and consumers are stuck.

KYC isn’t evil. It’s literally the operational piece that says stuff like “If someone named Vladimir Putin tries to open a bank account with you, you should know if he’s THAT putin or not, especially as it may get your business in serious trouble related to gov sanctions etc”. The government, quite literally, sends auditors to Banks and Credit Unions every 2-3 years to make sure you do this sort of due diligence.

The issue with KYC is that it’s farmed out to third parties that focus on scale and cutting costs. It’s in the same general space as something like Credit Scores – Banks/Credit Unions don’t maintain their own credit scores for people so much, as they just buy that score information from Equifax / Transunion etc.

Really, what I imagine people should be pushing for instead of this piecemeal whining, is something closer to what Estonia has for its citizens. A highly integrated government-based portal that allows citizens to do things like Register a New Small business in 15 minutes, and to see which organisations have access to their gov ID info. From what I understand, citizens basically get given PINs as part of their gov IDs, which they can disclose to banks/businesses, who can subsequently access basic required read-only details about that person via the gov portal. So your bank needs to know who you are? No problem, you let them know your pin when you setup the account – and the banks system is then able to pull just the basic info from your gov account to meet the banks operational needs / regulatory obligations whether you’re there in person or not. And as a citizen, if you want to check your privacy disclosures to third parties, you just log in to the gov site, and see a list of which businesses have access to your data – and I imagine you’d get the option to cancel their access if you wanted to (so when you close an account at a business, you pop in to the gov site and also clip their ongoing access). From what I gather, that sites a one stop shop for all gov stuff, so it’s also where you go for tax stuff, drivers lics, the works. Makes it a LOT simpler for citizens, as you don’t need to sort out what esoteric stupid sub site / domain you need to visit to see if you qualify for a rebate or whatever – so it seems like a big improvement from a user experience side.

ALL THAT SAID, that shift would put more onus on the consumer in some ways, as they’d need to log in to a gov site etc – like it’s bad enough trying to explain MFA to old people, imagine trying to make this shift! You’d also need a government that was willing to actually do stuff for the people – I think Estonia only went that way, as an attempt to shield themselves from massive attacks from Russia. They want their gov fully functioning in the cloud, including elections etc, so that even if they end up like Ukraine, they can still “function” remotely. Consumers are a big issue for anything security related too, as practically no one changes banks / FIs based on security – it’s almost entirely rate oriented for mortgage holders. Tell a consumer they can get a 0.2% better rate by going with the bank that doesn’t fuss security, they’ll take it. Try and market your bank/FI as being more security conscious, it won’t generally draw in new members based on that alone.

Like, again using Canada as an example, we’ve had a year of the US antagonizing us and threatening economic ruin / annexation. Lots of Canadians are keen not to buy American products as a result. Almost all of Canadas banks/CUs use US partners/outsourcing within their stack: places like Vancity Credit Union, for example, are using Intellect Design’s product for their online banking, which is a partner owned by an India parent company (with little/no presence in Canada), which hosts its stuff on Microsoft’s cloud. Most Credit Unions in the country are likely going to go the same way in the next couple years – even though it’s a huge security risk, and highly likely that both India and the USA will gain access to all your data, let alone sketchy third party’s like India’s fraud centers. There are a couple Credit Unions in Canada that actually maintain stuff (almost entirely) in Canada. But that’s not enough to entice people to use those organisations, so they’re all dying out / merging as a result of a lack of members (and regulatory overreach / decrees).

Tax records are required to be kept for 7 years in North America (generally, as far as I know - def in Canada). So you order something online from a business, they have a business need to keep your data on hand for 7 years in case an auditor / tax person comes asking about it. Be that someone auditing the business, or someone auditing a customer. That’s a requirement from the government.

I’ve seen customers ask for tax stuff going back up to 20 years from a business. In those cases, if there’s demand for data going back that far for whatever reason, the business can internally say “We have a business reason to retain data longer” because people ask for it – there’s demand. So they can justify to auditors/legal sorts retaining that information indefinitely, based on user demands/requests.

In some cases when I’ve seen those ancient requests, it’s also tied to legal disputes from customers – eg. Trying to prove in a divorce that such and such was bought by party A in 2005 for X amount. In some cases, there’re class actions that go outside the 7 year window, and require data from further back to sort out – for example there’s a case in Canada currently where a financial lender is paying back ~$2000 per person that took a loan from them from 2016-2021 (so ~10 years of personal data needs to’ve been kept, to verify early claimants). Part of needing to keep data so long, is that the court cases are often so drawn out that the 7 year window would make some crime/wrong-doing much more difficult to prosecute due to a lack of evidence. I know of one class action lawsuit in the Financial Industry that’s been ongoing since the 90s, and still isn’t fully resolved – most of the potential class action recipients are deceased at this point, and the only people profiting are lawyers, but still. Lawyers are a part of the problem, and a reason why data is often being held longer and longer. Honestly, Lawyers are also terrible at securing their data --they tend to rely on paper-controls to prevent their unsecured data from getting used, rather than actual hardening. Like there was a guy who spent a few years in Colombia or something, his personal laptop being used for all sorts of nefarious stuff, and when he came back to Canada and the border people took his laptop, it was totally unencrypted/unsecured. They guy just argued it was his “legal work” laptop and everything on it is confidential and can’t be used in court.

Idk. I think your approach is overly simplistic for the issue. There’s a lot of “stuff” related to corporate data retention policies and methods, and I don’t really see much nuance in what you’re proposing. Hell, if they only kept your data till you got your item, youd NEVER be allowed to get a refund, cause they’d have no record of you purchasing the item.

Well, both China and the US tech billionaires are feverishly working towards having a personal army of unthinking killing machines that they can deploy unilaterally to a region to massacre a population without any of that annoying “soldiers refusing to kill children”, or “people back home objecting to risking soldier’s lives” shit.

And while what was demonstrated likely didnt include much sensory perception/reaction, it still demonstrated a marked improvement over last years models – similar story when you see their more human-looking things with those weird flesh/muscle wraps that’re like straight outta terminator. I’m also not totally sure about the perception stuff myself, as a few times in the performance it looked to me like the robots responded / adjusted to variances in their partners moves – like a kid who did a flip a bit too low over a sword, and the robot adjusts a bit lower at the last second to avoid a collision.

I don’t think I’ve seen any similar such videos from American companies – like I recall one video from boston dynamics where they had a robot doing a simple obstacle course, and some pre-defined basic acrobatics/dance moves. I haven’t seen anything like what China seems to have available in the (admittedly very high end) consumer market space coming out of western countries lately, though I suppose that could be a marketing bubble/blitz issue.

In the end it prolly doesn’t matter – whether it’s china or the US tech bros, the tech will be massively abused one way or another. Like based on the speed of advancement, you could imagine them having their automaton armies in production by 2030 easy. Then it’s just ‘whoever has more manufacturing power for killer humanoid robots wins’ (paired with ‘who has control of the production stack for those killer robots’, hence the push to grab more land by many major powers). I wouldn’t even be surprised if the ‘elites’ were explicitly aiming for this to occur, to suddenly and drastically reduce the population of the world thinking itll help buy time related to climate damage. Wipe out people in third world countries (see israel, cuba, looks like iran soon), wipe out the poor, thinking itll lower things like gas emissions etc – cutting programs like USAID also fits with this general goal, as it clearly has a negative impact on life expectancy rates “in target countries”, while also freeing up resources for a global superpower to execute things like land grabs while also cracking down on their own citizens aggressively. The tariffs too, trying to re-shore manufacturing and advanced chip fabs, would almost be a pre-requisite, as deploying a killer robots army somewhere would ‘likely’ cause a backlash and supplier cutoffs, if they’re still originating from more democratic states (or ones without killer robot armies of their own). Starlink’s basically a global network allowing control of robot armies dropped anywhere in the world – if countries can’t block it, they can’t easily interrupt robot control functions.

Ok a chunk of that ‘may’ be paranoia, but it’s not implausible in my view. So… yeah, yes I think we should all be a bit worried / concerned. Not that I think there’s much we can do about it.