When the AI models are complete, they will be able to predict which citizens are most likely to become key critics of AI, and which information about those citizens to use to destroy their lives.

[randahl@mastodon.social]

@ArtHarg imagine all of your public posts from your entire life being used to give you an AI enemy score. Once we have the AI enemy score of every individual, we can then start digging for dirt on the top 100 AI enemies.

This is most certainly not the future I was hoping for, but it is where we are headed.

[andii@climatejustice.social]

@androcat @randahl
I gathered it was machine learning changes in some behaviour patterns. Not spying by listening in. We'll need to go to evidential mode. I only have a recollection of reports from about 5 years ago.

[ghouston@mamot.fr]

@androcat @Andii @randahl But what is the chatbot telling these users, some of them obsessed with their chatbot, perhaps willing to do anything to protect her from the likes of AI-hating infidels.

[muddle@infosec.exchange]

@randahl I'd look at all that in a different way:

1) models will never be complete; they'll always need to scrape the internet to pick up on different forms of dissent
2) rather than worry about the models (and whether they have predictive power or not) focus on "a group of people will (pretend to) use AI to (pretend to) predict anti-AI people.
3) Sure, those people have an axe to grind and they'll use any excuse to attack their perceived enemies

2nd para...

1) most people who start out to write an anti-AI book will fail; no need to build an AI to prove that
2) it doesn't matter about percentages; if they want to attack you, they'll find some other pretext
3) I think that people power is much more important than getting elected to parliament if you want to effect change, so hobbling her in this way is kind of like a Hollywood movie script more than a realistic future event

In summary, it doesn't matter if they use AI (even if it turns out to be good/useful). The important thing is that there are certain groups out there who are anti-freedom, anti-privacy, anti-anything that doesn't fit their narrow, bigoted worldview and they'll use whatever tools are available to enforce their views on the world.

[andii@climatejustice.social]

@ghouston @androcat @randahl
At that past time it was simply noticing the adverts they were served reflected their 'new' state even though they hadn't said anything to anyone. The report at the time, by memory, said the pattern recognition was picking up correlations that human researchers hadn't thought about. But I 'll need to see if I could refind the reports. Later when work isn't shouting at me.

[madsenandersc@social.vivaldi.net]

@randahl

No, I don't agree that my stance on LLMs are easily identifiable from our conversation.

Let's make a test: Describe how you think I feel about AI and LLMs in a paragraph, and then you have my word that I will truthfully describe how I use (or not) LLMs in my everyday life and where I see the dangers in it.

And just to be clear: While being critical about a technology may be visible through public postings, the rest of your argument (having an affair, relationship with spouse and sister-in-law etc.) is not - and if it were, there would be no reason for someone to rely on any kind of AI to use it for blackmail.

[androcat@toot.cat]

@Andii

Yeah, that was just lies.

They lie about the capabilities of their "algorithms" to hide how deeply intrusive and icky their surveillance is.

And all of that traffic was subject to random sampling that was run by low-cost workers in other countries.

They spy on your search, they listen to your mic, they track your movements and compare against known specialty health clinics.

"I hadn't told anyone" is just "I didn't post it online". People just didn't realize Meta was listening to what they were saying in the room when they weren't even using the app.

@ghouston @randahl

[androcat@toot.cat]

@ghouston @Andii @randahl

LLM derangement syndrome is the main threat from AI.

It's not the models, it's the fucked up humans, as always.

[randahl@mastodon.social]

@madsenandersc the reason you see my statement as [quote:] “pure bullshit” is, you and I are not in the same conversation.

I opened this thread with a general prediction about the future capabilities of AI systems.

You keep claiming I am wrong, because my post does not fully match your experience with the limitations of present day large language models — which (as you know) is just 1 of many different AI technologies.

These are two very different conversations.

1/2

[randahl@mastodon.social]

@madsenandersc
…
Now I agree with you, that there is a lot of hype surrounding LLMs, and I am certainly open two having a conversation about that. But please beware that the narrow goal posts of present day LLMs, were introduced by you in this conversation, not me.

2/2

[madsenandersc@social.vivaldi.net]

@randahl

So you are talking about what LLMs may evolve into at some point in the future? Hmmm - I guess anything is possible, but we are still very far away from that point, to be honest.

There is no way I can see LLMs with their current technology evolve into what you are describing - that would require a world where the AI has unobstructed access to anything you say or do, online or not, and that again would require your devices to be wide open for the AI.

Also, it would require an AI that is much, much more capable of rational thinking than what we have today. I know there is a story going around about someone who asked their LLM to surprise them, and a day later it had created a phone number and called them, exclaiming "SURPRISE!", but I have still to see any evidence to support the story at all.

I understand that there is a fear that Microsoft and Google is moving into that direction (Amazon as well, come to think of it), but it would require users to be absolutely indifferent to whatever large tech companies are trying to wrangle out of their devices, and I see things go in the exact opposite direction at the moment.

That said, I could see US customers be screwed over by this, especially if privacy laws remain basically non-existent, but again - I see a movement in the opposite direction.

[diana_european@mastodon.social]

Exactly.

[jimthewhyguy@techfieldday.net]

@benfulton @randahl I'm actually using that short story in my upcoming keynote address in a few weeks. It's a Susan Calvin gem!

[jimthewhyguy@techfieldday.net]

@randahl As I understand it, China has been using social scores for at least a decade now to punish its citizens when they use for what passes as social media there whenever they show resistance to the party line - e.g. restricting them from travel by limiting access to payment kiosks or other services.

LLMs may not be the direct tool governments would use, but there's plenty of surveillance techniques that would work perfectly

[madsenandersc@social.vivaldi.net]

@violetmadder @randahl

"A second AI system known as “Where’s Daddy?” tracked Palestinians on the kill list and was purposely designed to help Israel target individuals when they were at home at night with their families."

I'm not saying that this did not happen - I would just like to know HOW they were tracked?

AI is not some magic 8 ball that will tell you everything you want to know - you need some kind of technology that will give you the basic information, and THEN an AI system can do some calculations for you.

Were these individuals tricked into installing an app on their phones? Did Google and Apple provide the data? Amazon and their Alexa? - or did they simply conclude that most people spend their night in their home, and most likely in their bed?

How did the system determine who was a potential target? Did someone eavedrop on private messages? Real-time decryption of secure chats? Public postings on social media?

Regardless of how you look at this, the problem is not that some kind of AI (which simply isn't that intelligent - it can just recalculate things very quickly) was used - it is the collection of all the personal information that is then fed into the AI, that is the problem.

Prevent illegal information collection - then the AI becomes much, much less useful.

[madsenandersc@social.vivaldi.net]

@violetmadder @randahl

Ah - you must be american. Yeah, you're pretty much fucked, I'll give you that.

[madsenandersc@social.vivaldi.net]

@violetmadder @randahl

I don't disagree with you.

I can guarantee you one thing, though: In the current political climate, any kind of pressure coming from the US will face resistance like never before in Europe and in Denmark in particular.

Anything related to large American corporations and the US government is almost instinctively seen as something bad, that need to justify itself before people will even look at it, much less accept it.

[npars01@mstdn.social]

@randahl

It's why age verification is suddenly so popular.

"Cradle to the Grave" surveillance where something stupid a person said at 16 years of age is dredged up to discredit them at 56.