IDK how we solve this without some serious regulatory changes and the backing awareness and political will.
An “a scanner darkly” approach could be kind of effective to pollute online information about yourself.
If there are 1000 “me” profiles all saying different stuff, it’d be somewhat hard to pin down what I think and persecute me. This doesn’t work as well or maybe at all for sufficiently doxxed people who are producing content for a living, public political organizers, etc.
> If there are 1000 “me” profiles all saying different stuff, it’d be somewhat hard to pin down what I think and persecute me.
Noise-drowning only works against profiling that's set up with intent to measure you, rather than punish you.
I'm afraid, in autocracies and dictatorships, it takes one "me" profile (that could be not even mine) that said something out of current goodthink to be persecuted. Just look at what's going on in Russia.
The key point of those "we'll be using AI to monitor everything you publish" announcements is not in even it its directly stated goal. It's in the chilling effects this creates, making people avoid expressing their opinions except for unquestionably safe ones.
The same would happen in the EU, where police forces are scanning social media to find offenders.
I'd strongly advise having no public profile of any kind, or perhaps only a professional one where you religiously follow the rule of not posting anything remotely private.
Why do you think regulation will help? AI is regulated in Europe ... and this is happening anyway. You see, all the EU regulations, going all the way back from the GPDR, have provisions making their intention very clear:
1) A number of actors can (and do) make exceptions to the law for anyone and everyone they want. These actors include all member governments and a few others, like ChatControl's Europol, and the EU commission.
2) Individuals do not have standing to sue under these laws. Only executive governments do. If AI is used against you, or your privacy is violated, you get to beg the government to sue (not even any parliament or court, the executive, ie. "the minister"). You can also not get damages or injunctions under these laws.
3) the penalties if you get convicted are absurdly large, right from the first offense. Large enough to destroy most companies in one go. It's "death penalty for wrongly tied shoelaces" approach, very selectively enforced.
This is why you're seeing the behavior in the market in the EU that you're seeing now. A bunch of actors have the protection of the government, and just violate these laws on an incredible scale. On the other hand, these laws are being used as weapons against, for example, the FANG companies. And everyone else is scared.
Governments in the EU (and ...) can and do use mass surveillance for, frankly, stupid purposes (recently I noticed parking tickets in a bunch of regions of Prague work by using mass surveillance to get license plates with many cameras through AI. And while I admire the technical part of it ...)
So AI is regulated in the EU, in the sense that nobody in the EU gets to use it against a state's wish. This has been done incredibly heavy handed. From the perspective of citizens however, it is effectively not regulated. You are being face recognized in all large cities. All license plates are being scanned with overviews of who is where, live reporting on cellphone activities for how people coordinating (as in which phone app they're using for this protest, whatsapp or signal?) available to the police live. Your medical data is being scanned by insurance to deny coverage (because that's an allowed exception), or for example, to direct youth services to your kids if you get psychological treatment. Any psychological treatments are available to the major and judges in cases of either family court or to declare you legally incompetent or ...
Since it's not integrated across countries you can, for the moment, escape this by getting treatment in a neighboring country and paying cash. You see, EU governments don't trust each other so they only share data in specific cases, which have to be approved in most situations (funny how governments don't trust anyone else, not even their closest allies, with this data)
Most people are fine w/ the deal corporations like Alphabet & Meta have made w/ them. They are perfectly happy to trade their privacy for convenient access to "information" held on their servers & data centers. I don't think AI makes much difference here b/c if we can trust the numbers from OpenAI & Google then it seems like people are happy w/ the deal here as well. They don't mind letting these corporations gather their queries & feedback as long as the perceived value they get from the software justifies paying their subscription fee. Politicians understand all this & so they've determined that the people are ok w/ any amount of surveillance as long as they get some amount of utility out of it so they will continue to build out the data centers & surveillance infrastructure. It's not like there aren't satellites already in orbit that can zoom in on a penny in a parking lot anyway so what's a few more data centers & cameras to add to the mix & then pipe the data stream into an AI for "anomaly" detection?
That's why governments do it using face scanning cameras, license plate scanning, downloading social media profiles, uploading all medical data to government servers (minus any large diagnostic data, like images, because that costs too much), demanding FANGs give them access to chats, ... You know, behind people's backs.
Because if figure out what is done with this data, who has access to it, people would never agree to that. For example, youth services uses data on medical treatments people in families receive to find kids to "protect" (ie. abduct). Medical data is also shared with nationalized insurance and hospitals to deny treatments "if it's your own fault" (e.g. find past drug tests in case of organ transplants)
Btw: if you're in the EU, remember, in almost all cases you can refuse a drug test, or any test, and for the love of god, in all cases you want to refuse this. First: the fact that you had a drug test for any reason will be used against you. Second: if for whatever reason it's positive and you need a new heart 20 years later ... best of luck.
The way you know it's gone pretty far is that mass surveillance + AI is being used for parking tickets. How far down the priority list is that?
Imho, these are things people would never agree to if the choice was clearly presented in parliament. No way. And that's in democracies.
I read Gov Greg Abbott's book Broken but Unbowed -- I am generally not his supporter, but grew up among semi-conservative Texans under Governor Ann Richards (D) — loved her wit (she was also a family friend, Beloved Woman)!
An (the only) interesting concept that Gov Abbott (R) explores in his book is about the US States' collective right (under our Constitution) to call for a national convention for ANY REASON, initiated by a super-majority of states in mutual agreement on topic only, in order to propose Amendments (all prior were done by the Federal Congressional route).
This seems like the most-reasonable solution to introduce some sort of technology bill of rights (e.g. privacy; AI governance; data collection practices; right to cancel; opt-ins) — to address limitations to our geriatric Congress' inability to get with the goddamn modern times. We still operate on telecommunications laws from the friggin' eighties!!!
We might as well overhaul all of healthcare, too... and this really isn't too far fetched (I don't have the book in front of me... but there have been prior attempts to call such state conventions, decades ago).
...perhaps end Citizens' United? It's certainly time for some citizen initiative (via our State Congresses).
----
If Sir Glubb's Fate of Empires taught me anything, most societial governances don't last too much longer than 250 years (¡happy birthday USA!). I think ours can, with massive but only with massive overhaul.
Giving advice to Gov Abbott: you're doing a great job for your state as Governor; stay there (don't run for president, again — you'll just waste time better-spent preparing Texas for what's next?); perhaps you should lift the boot up off your citizens' impoverished throats just. a. little. bit. ...but otherwise, Keep Texas Beautiful. [and say `hi` to James for me — it's been too long].
Instead of making this Republican vs Democrat issue, how about you complete that paragraph.
"Suresh served in the Biden-Harris administration as Assistant Director for Science and Justice in the White House Office of Science and Technology Policy. In that capacity, he helped co-author the Blueprint for an AI Bill of Rights. He was named by Fast Company in 2023 to their AI20 list of thinkers shaping the world of generative AI."
Seems to me he wants rules and legislation around AI so it protects the American people. AI Bill of rights was deleted when this new administration took over but available at https://archive.is/XEFJJ
"To advance President Biden’s vision, the White House Office of Science and Technology Policy has identified five principles that should guide the design, use, and deployment of automated systems to protect the American public in the age of artificial intelligence. The Blueprint for an AI Bill of Rights is a guide for a society that protects all people from these threats—and uses technologies in ways that reinforce our highest values"
Yeah .. no . Firstly there’s really no such thing as “surveillance capitalism.” If we’re going to start coining terms like that, then why stop there .. lets call Oracle and Postgres “database capitalism or Microsoft vs Apple as “OS capitalism .. or Salesforce “CRM capitalism,” or even Google Docs as spreadsheet capitalism .. anyone who writes a an article with this sort of title is usually (actually almost always) a fraud
That's your opinion and that's fine but surveillance capitalism is not a freshly minted term - it simply supposes that the behavioral data collected on all of us is a new form of capital to be accumulated, exploited, and hoarded
Yes, that I agree with .. but it’s still a pointless term crafted more for optics than substance. It doesn’t really add clarity, it just feels like linguistic inflation. Same vibe as when people say “weaponisation” to inject drama and a false sense of danger
Man up and say what you want to say. None of this weakling passive aggressive quoting a line from the article like we're supposed to believe that makes it a smoking gun.
If you think it's a defensible position that the only people who can have a legitmate opinion on this topic are the hypothetical pure neutrals who don't affiliate with any party, make it yourself and stand by it.
IDK how we solve this without some serious regulatory changes and the backing awareness and political will.
An “a scanner darkly” approach could be kind of effective to pollute online information about yourself.
If there are 1000 “me” profiles all saying different stuff, it’d be somewhat hard to pin down what I think and persecute me. This doesn’t work as well or maybe at all for sufficiently doxxed people who are producing content for a living, public political organizers, etc.
> If there are 1000 “me” profiles all saying different stuff, it’d be somewhat hard to pin down what I think and persecute me.
Noise-drowning only works against profiling that's set up with intent to measure you, rather than punish you.
I'm afraid, in autocracies and dictatorships, it takes one "me" profile (that could be not even mine) that said something out of current goodthink to be persecuted. Just look at what's going on in Russia.
The key point of those "we'll be using AI to monitor everything you publish" announcements is not in even it its directly stated goal. It's in the chilling effects this creates, making people avoid expressing their opinions except for unquestionably safe ones.
The same would happen in the EU, where police forces are scanning social media to find offenders.
I'd strongly advise having no public profile of any kind, or perhaps only a professional one where you religiously follow the rule of not posting anything remotely private.
Why do you think regulation will help? AI is regulated in Europe ... and this is happening anyway. You see, all the EU regulations, going all the way back from the GPDR, have provisions making their intention very clear:
1) A number of actors can (and do) make exceptions to the law for anyone and everyone they want. These actors include all member governments and a few others, like ChatControl's Europol, and the EU commission.
2) Individuals do not have standing to sue under these laws. Only executive governments do. If AI is used against you, or your privacy is violated, you get to beg the government to sue (not even any parliament or court, the executive, ie. "the minister"). You can also not get damages or injunctions under these laws.
3) the penalties if you get convicted are absurdly large, right from the first offense. Large enough to destroy most companies in one go. It's "death penalty for wrongly tied shoelaces" approach, very selectively enforced.
This is why you're seeing the behavior in the market in the EU that you're seeing now. A bunch of actors have the protection of the government, and just violate these laws on an incredible scale. On the other hand, these laws are being used as weapons against, for example, the FANG companies. And everyone else is scared.
Governments in the EU (and ...) can and do use mass surveillance for, frankly, stupid purposes (recently I noticed parking tickets in a bunch of regions of Prague work by using mass surveillance to get license plates with many cameras through AI. And while I admire the technical part of it ...)
So AI is regulated in the EU, in the sense that nobody in the EU gets to use it against a state's wish. This has been done incredibly heavy handed. From the perspective of citizens however, it is effectively not regulated. You are being face recognized in all large cities. All license plates are being scanned with overviews of who is where, live reporting on cellphone activities for how people coordinating (as in which phone app they're using for this protest, whatsapp or signal?) available to the police live. Your medical data is being scanned by insurance to deny coverage (because that's an allowed exception), or for example, to direct youth services to your kids if you get psychological treatment. Any psychological treatments are available to the major and judges in cases of either family court or to declare you legally incompetent or ...
Since it's not integrated across countries you can, for the moment, escape this by getting treatment in a neighboring country and paying cash. You see, EU governments don't trust each other so they only share data in specific cases, which have to be approved in most situations (funny how governments don't trust anyone else, not even their closest allies, with this data)
Most people are fine w/ the deal corporations like Alphabet & Meta have made w/ them. They are perfectly happy to trade their privacy for convenient access to "information" held on their servers & data centers. I don't think AI makes much difference here b/c if we can trust the numbers from OpenAI & Google then it seems like people are happy w/ the deal here as well. They don't mind letting these corporations gather their queries & feedback as long as the perceived value they get from the software justifies paying their subscription fee. Politicians understand all this & so they've determined that the people are ok w/ any amount of surveillance as long as they get some amount of utility out of it so they will continue to build out the data centers & surveillance infrastructure. It's not like there aren't satellites already in orbit that can zoom in on a penny in a parking lot anyway so what's a few more data centers & cameras to add to the mix & then pipe the data stream into an AI for "anomaly" detection?
That's why governments do it using face scanning cameras, license plate scanning, downloading social media profiles, uploading all medical data to government servers (minus any large diagnostic data, like images, because that costs too much), demanding FANGs give them access to chats, ... You know, behind people's backs.
Because if figure out what is done with this data, who has access to it, people would never agree to that. For example, youth services uses data on medical treatments people in families receive to find kids to "protect" (ie. abduct). Medical data is also shared with nationalized insurance and hospitals to deny treatments "if it's your own fault" (e.g. find past drug tests in case of organ transplants)
Btw: if you're in the EU, remember, in almost all cases you can refuse a drug test, or any test, and for the love of god, in all cases you want to refuse this. First: the fact that you had a drug test for any reason will be used against you. Second: if for whatever reason it's positive and you need a new heart 20 years later ... best of luck.
The way you know it's gone pretty far is that mass surveillance + AI is being used for parking tickets. How far down the priority list is that?
Imho, these are things people would never agree to if the choice was clearly presented in parliament. No way. And that's in democracies.
I've only gotten 1/4 through, but isn't the horse thoroughly dead at this point...
privacy laws and $25/hr min wage would change everything right now
I read Gov Greg Abbott's book Broken but Unbowed -- I am generally not his supporter, but grew up among semi-conservative Texans under Governor Ann Richards (D) — loved her wit (she was also a family friend, Beloved Woman)!
An (the only) interesting concept that Gov Abbott (R) explores in his book is about the US States' collective right (under our Constitution) to call for a national convention for ANY REASON, initiated by a super-majority of states in mutual agreement on topic only, in order to propose Amendments (all prior were done by the Federal Congressional route).
This seems like the most-reasonable solution to introduce some sort of technology bill of rights (e.g. privacy; AI governance; data collection practices; right to cancel; opt-ins) — to address limitations to our geriatric Congress' inability to get with the goddamn modern times. We still operate on telecommunications laws from the friggin' eighties!!!
We might as well overhaul all of healthcare, too... and this really isn't too far fetched (I don't have the book in front of me... but there have been prior attempts to call such state conventions, decades ago).
...perhaps end Citizens' United? It's certainly time for some citizen initiative (via our State Congresses).
----
If Sir Glubb's Fate of Empires taught me anything, most societial governances don't last too much longer than 250 years (¡happy birthday USA!). I think ours can, with massive but only with massive overhaul.
Giving advice to Gov Abbott: you're doing a great job for your state as Governor; stay there (don't run for president, again — you'll just waste time better-spent preparing Texas for what's next?); perhaps you should lift the boot up off your citizens' impoverished throats just. a. little. bit. ...but otherwise, Keep Texas Beautiful. [and say `hi` to James for me — it's been too long].
[flagged]
[flagged]
Instead of making this Republican vs Democrat issue, how about you complete that paragraph.
"Suresh served in the Biden-Harris administration as Assistant Director for Science and Justice in the White House Office of Science and Technology Policy. In that capacity, he helped co-author the Blueprint for an AI Bill of Rights. He was named by Fast Company in 2023 to their AI20 list of thinkers shaping the world of generative AI."
Seems to me he wants rules and legislation around AI so it protects the American people. AI Bill of rights was deleted when this new administration took over but available at https://archive.is/XEFJJ
"To advance President Biden’s vision, the White House Office of Science and Technology Policy has identified five principles that should guide the design, use, and deployment of automated systems to protect the American public in the age of artificial intelligence. The Blueprint for an AI Bill of Rights is a guide for a society that protects all people from these threats—and uses technologies in ways that reinforce our highest values"
Yeah .. no . Firstly there’s really no such thing as “surveillance capitalism.” If we’re going to start coining terms like that, then why stop there .. lets call Oracle and Postgres “database capitalism or Microsoft vs Apple as “OS capitalism .. or Salesforce “CRM capitalism,” or even Google Docs as spreadsheet capitalism .. anyone who writes a an article with this sort of title is usually (actually almost always) a fraud
That's your opinion and that's fine but surveillance capitalism is not a freshly minted term - it simply supposes that the behavioral data collected on all of us is a new form of capital to be accumulated, exploited, and hoarded
Yes, that I agree with .. but it’s still a pointless term crafted more for optics than substance. It doesn’t really add clarity, it just feels like linguistic inflation. Same vibe as when people say “weaponisation” to inject drama and a false sense of danger
Man up and say what you want to say. None of this weakling passive aggressive quoting a line from the article like we're supposed to believe that makes it a smoking gun.
If you think it's a defensible position that the only people who can have a legitmate opinion on this topic are the hypothetical pure neutrals who don't affiliate with any party, make it yourself and stand by it.
[flagged]