Thanks Bettina. Yes. De growth seems to be the ultimate aim of all this. It certainly seems the likely consequence.
I’m just not sure where to take all this tbh. I’m spending a lot of time on Substack as I get less busy elsewhere. But it doesn’t really offer much of a (financial) return in terms of time invested.
Merry Christmas to you Bettina. And thank you so much for your continued support. It is genuinely very much appreciated.
AI got the image a bit wrong - TwoTier is a lot chunkier than that.
Funny how TPTB in the UK want to take us back to a pre-science medieval life, without the religion (except Woke) and without electricity and - oh yes - real feudalism. Back in your box peasant!
Merry Christmas to you, LSO, and Happy New Year - I hope the 'crossroads' doesn't take your Substack away!
I wholly agree with you that "if it saves just one life" is about the worst reason possible for banning anything. We would have no food, shelter, power or transport.
Normally the issue with technology is overspecification. The camera on my phone would have beaten most of the kit David Bailey used in his heyday, but the photos don't. Partly because they are of things like hard to reach consumer units so I can read the small label telling me which MCB turns what things off. I don't do cat videos.
The reverse is true with AI at present, because it isn't intelligence at all. It doesn't have the capacity to understand why the question is asked, and so it just regurgitates stuff from the internet, and makes up things to fill the gaps. You can't beat it for telling you how to write a complex formula in Excel, as long as you know how to ask the question, but ask it to write a short story with Poirot as the main character in the style of Agatha Christie, and the result is pathetic. This is probably why those studying the impact of feminism on glaciology at Nopoint University get AI to write their garbage, and their "professors" get AI to mark the AI.
Frankly, AI today is not much more than a slightly better search engine for specific things, or yet another endless time waster like computer games and cat videos.
Of course, it may one day become intelligent, at which time we will have to dust off those old Terminator DVDs to work out how to deal with it.
Absolutely Jeremy. And yet they always use it as justification for the most draconian restrictions. I heard the ‘just one life’ excuse endlessly during lockdowns. It was incredibly disingenuous.
The idea that a LLM could become sentient is equally bonkers. Sure they can be impressive as you say, though only if you plug in the ‘correct’ inputs. The idea that they somehow ‘understand’ what any of the terms are is nonsense. They are just a’ Chinese Room’ if you know that idea.
I have to say though, I love video games and cat videos. Two of my very favourite time wasters.😆
Not often I have reservations about the conclusions and observations that you draw, LSO, but I could not be more cautious and pensive about the inexorable march of AI.
Here we are, humans, possessed of arguably the finest intelligence in history (if we set to one side that of dolphins and orangutans, both perfectly in balance with their surroundings and essentially kindly in their nature) and we are seemingly content to let the thumb-enriched players of computers to dictate how matters are to be developed in the future.
AI cannot scent the rain, nor see the love in a dog's eyes ... I tend to stick the to the real and the stuff not dependent on the click of a computer - but then I am very old and rather old-fashioned, and therefore not entirely to be trusted on matters in this new age.
I’m delighted to have this discussion with you no-one. I have a huge respect for your position. It is shared by many people. Not least Elon Musk. Who seems to know a thing or two about this stuff.
However I worry that the threat of AI, at least at its current level, is being massively talked up by governments as an excuse to censor and monitor adults. It is the perfect labour saving tool of authoritarians.
But too dangerous to allow into the hands of little people.
I can see ‘Have you not seen Terminator?’ Becoming the new ‘Won’t you think of the children?.’
But certainly this is a discussion we should be having. And thanks for having it with me.
Indeed, LSO. Digital impulses may be capable of many things but when I read of the electronic, blisteringly fast cogitations of AI, and the essentially plagiaristic nature of its pronouncements, I wonder if it could ever truly understand the humanity of someone like Alexander Pope:
I’d never heard of the Carrington Event before. I’m going to look into that some more. Fascinating.
My dad could put a gearbox in a car and plumb and tile a bathroom from scratch. I have barely any of those skills yet I’m practically an engineer compared to my peer group.
The skills you have been pretty much eradicated from our culture. I believe for practical purposes we are all cyborgs now. Only capable of sustaining ourselves with the aid of machines.
Another ‘event’ and I’d only survive until my dried fruit and nuts run out. About a week if I was careful and didn’t share any with my kids …🤣
With regard to the plagiaristic nature of AI art I think Pope would immediately see the danger of all this then.
That currently AI is feeding/learning from human created art. But it will reach a point when most of the ‘art’ online will be AI generated. At that point it will become an ouroboros, forever eating a regurgitating its own AI tail. Weird. What will become of us then?
I’m intrigued as I am worried to be honest. But maybe I’m just being foolish.
Never foolish, LSO. All electronic 'stuff' is vulnerable eventually, if not to the simple 'off-switch' then to a Carrington Event (last witnessed in 1859 and long overdue a repeat, apparently. Ref: https://en.wikipedia.org/wiki/Carrington_Event)
For myself, having always been a bit of an an outdoors chap, so to speak, I have made sure that my children and grandchildren can forage for food, light a fire in the damp early morning, find water where there appears to be none, climb a cliff without ropes, ride a horse, and put an arrow into the gold at fifty yards. Call it neanderthal primitivism if you will, but I suspect it will serve them better in the near to medium term than any amount of button-pressing as we venture into the coming collectivist nirvana.
AI is an interesting one: for folk like me it generates 90% of my code and then I fill in the complicated bits - by asking it more questions. It also explains what it's doing and I comment throughout and look really clever, which I'm not.
AI for government and statist organisations is an absolute terror, so no wonder the EU and Left have banned it. Imagine a genuine AI system operating solely from logic, collating thousands of data points per millions of folk in seconds (doing away with the department that is suppose to do it) and simply saying 'here's your parking ticket at a football game. Here's you passing through the turnstile. Here's your gym membership and you go to this class regularly. Your welfare payments will be stopped'.
Or, the real terror for big government: "this department does not need 20,000,000 quid. Forty percent of those here do nothing of value. The department budget will be reduced by 70% as the 50 quangos and charities who demand money of it do not need to exist either."
Imagine AI's opinion of the EU! A pointless talking shop that has hindered economic growth, prosperity and freedom for no benefit but that of those inside the thing.
We're a long, long way from true, aware thinking AI but considering where we are and where we will be and government should be afraid. As it is we are witnessing the last flailing abuses of a system so self serving, arrogant and utterly terrified of losing the control it is frantically giving what little relevance it has away to a body utterly untouchable by the public.
Great points Wibbling. I’m pretty sure that the bosses are aware of them all. Which is why they won’t take their eye off the ball like they did in the early days of the internet, allowing it to become, for a while, a proper free for all, which released all the innovation and groundbreaking services we see today.
Services like Google maps. YouTube. Paypal.plus social media sites like Facebook would have been throttled at birth if the regulators had been as in charge then, as they are now.
The idea is clearly to take charge of AI at this current level. And use it to police our online communications and limit our interactions. Flagging ‘problematic’ content and deleting any form of dissent it before it can infect the minds of free thinkers and sceptics.
Big fat state will try. It'll try legislation, oppression, suppression and when all that fails outright control; but it *will* fail.
Somewhere, someone will ignore their stupid dictat, maybe in another country and that'll be it. Government is like the courtiers for Canute, although rather than believing the king can hold back the tide, they think they can.
The state machine is wrong. Massive job losses, the WFA, high unemployment, inflation, debt, rich folk leaving, industry evaporating, expensive food, no growth (and the inevitable recession) are all results of sky high energy prices. Government is a cartoon character jamming fingers in the pipe it's drilling through and failing utterly to stop the problem they're causing. These people are dangerous, stupid, mendacious, spiteful, egotistical, greedy, incompetent, vicious, deceitful and arrogant. That is why everything they do fails. AI will be no different.
It seems clear to me that the biggest issue facing any independent-minded individuals today is the provenance of digital evidence. When AI can increasingly generate images and videos that can not easily be detected as generative, we know longer can know much of anything. I'm envisioning a world in which someone can, say, come up with a video clip that purports to be from the bodycam of a police officer doing and saying virtually anything the prompter wants. If we don't solve this problem, governments are going to use it as a way to control or shut down all the foundations of our independence.
To some extent this is already happening Jeff. Maybe not the creation of images but certainly the manipulation of data.
As you can probably see I work hard to link supporting or salient articles to my pieces. Google undoubtedly uses AI to downvote or ‘hide’ the ‘wrong’ sources.
If I look for anything vaguely controversial or a ‘conspiracy theory’ say Climate Change for example. Then a search for dissenting voices is usually fruitless unless you already have a source, like Spiked! for instance.
I have a lot of time for different opinions on this subject. But it is increasingly clear to me that the threat isn’t so much ‘AI’. It is ‘AI plus government’.
Google specifically excludes and 'deranks' (if such a word exists) sceptical arguments. For those of us wanting information the relentless promotion of political opinion merely demonstrates how little there is.
It definitely seems to from my perspective. I try and save as much as I can in terms of links when I first see a story. Because if I don’t at the time, I know it will prove v difficult to find them later.
I stopped using Google for search and everything else when it became clear they were manipulating it. It can be challenging. I use Presearch for search. It rarely disappoints. It's an open-source disributed search engine which funds server costs with minimal advertising. They've got some kind of crypto currancy reeard sysrem for viewing ads which I ignore. Try presearch.
Duckduckgo apparently doesnt do much native search themselves, but runs your search through Google but strips your identifiers so the latter can't track you. You still get Google's bias. Google likes DDG because it protects them from monopoly attack.
We've reached a point as a nation where double standards are rife and obvious, with no punshments for those abuses of justice and morality. In this world, truth is no longer a defence. The state doesn't need to fabricate evidence. Punishment depends on affiliation (for or against big government), not for actual crime.
I use AI generated stuff at work and refer to it as Artificial Stupidity it's that rubbish. It's only as good as it's programmed to be, garbage in etc. Totally agree re Poundshop Voldemort. He's a traitor to this country and should be dealt with accordingly. I can't express in words how much I despise him. I think there's a reason he's rarely in the UK since he took office a) he hates the UK b) he knows by now how much the population loathes him. We're all getting recession for Christmas and he may have concerns about what the fallout will be....
That’s really interesting Brendan. I hadn’t really considered that before, but of course, it’s a perfect fit. More stuff like this would be great. There’s so much potential, along with the inevitable catastrophising.
I happened to be at last years "British Machine Vision" conference (playing in the dance band after dinner) and got to watch some of the presentations and speak to some of the attendees, from all over, not just Britain.
"AI" is a broad generalisation for a collection of developments including neural networks etc. It was surprising and interesting to me as it probably is to you. That's an indication of how poorly the news media is serving us on this topic. Which is also a function of how technically illiterate they are.
Look up BMVC2023.org and you'll find the conference proceedings online for this and other years. Just scanning through the synopses of the papers will give you an idea of the breadth and depth being covered.
Not completely with you on this. US corporations have prioritised their profits over children’s lives when they have refused to resource changes to their social media algorithms which pull children to harmful and self-harm material. National governments need to bring home their responsibility to them.
Very fair point Jaffa. But governments won’t use these powers to protect children. Despite their protestations that this is their only motivation. Instead they will use these powers to censor, and to remove the rights and freedoms of adults.
We must embrace AI. If the sane among us shun it, the feral left will gain even more ground. Like a lot of things, it can be used for good and bad. It's a shame we don't have Sora. I am getting a bit fed up with using Bing AI for my work. They make some really good pictures, but too often there is a incongruous or bizarre thing thrown in that ruins the image.
I tend to agree Adam. We are at a junction in terms of technology. The best way to stop the government using AI against us is probably (though I’m not nailing my colours to the mast in this) making it available for everyone.
It’s a bit like the argument that Americans Second Amendment protects the First. Not sure I’m entirely convinced but even so it is a compelling argument.
I expect the US government will at some point in the next two years attempt to class some forms of advanced AI as ‘munitions’. That way it can set all the parameters for AI use that it wants.
This was exactly the argument David Brin made in his 2000-something book, "The Transparent Society". He went into the research for that book looking for reasons why we need strong guarantees of privacy. What he discovered while doing the research completely turned his thinking around. His conclusion was that if we try to regulate things like privacy (as the EU is currently doing) all that will happen is that governments and the rich and powerful will continue to use the technology that violates it, while the rest of us will be forbidden from being able to also use it to fight back. My paraphrase of his argument is grossly over-simplified. I recommend the book, even though it is old now.
We keep handing government more and more powers to ‘keep us safe’. But we end up no safer than we were at the beginning. We just have less power to do anything about it.
We don't exactly "hand" government more powers, they just keep taking them... whilst enough keep voting for the sort of government that will ensure the practice continues, at least in the UK. OTOH, they don't even provide the safety that most would value, such as protection from crime.
A Very Happy Xmas LSO and a more prosperous 2025!!!
The Quangocracy at full force. My opinion as it is, is that if body such as this oversteps its original mandate but is still considered necessary, the public gets to vote on who runs it and how much funding it gets. Directly. Because it has become a de-facto arm of the state.
We can also vote to disband it.
Most if not all of the functions can be handled simply by the government responding to citizens suggestions or complaints.
It’s a good plan Overhead. My worry with all this stuff is how easily it gets captured. Just thinking this morning about all the ‘independent’ institutions which aren’t. And how ‘devolving’ democracy to ‘the local level’ often leads to less democracy.
I think ultimately we must pare back the state significantly. Give it much much less to do, and relentlessly hold the people doing those things to account.
But yes. I’d take your plan over the current stitch up. ATB
Hi LSO, good piece. Generative AI is in its infancy. There’s going to be lots of dead ends, but when the dust settles, just like the dot.com era, it will change the way we work (already is in some sectors). That of course, is if there is much of an economy or for that matter country left after 5 years Labour to cap off 14 years of the Tories. As the Chinese curse would say, we live in interesting times.
Hi P. Yes there’s no putting the genie back in the bottle. for better or for worse. Who knows where this stuff will lead us? All I know is, if we let the government convince us it’s too dangerous for us, but OK for them, then we’ll end up nowhere good.
Merry Christmas to you and yours, and thanks for all your support in 2024. Much appreciated.
Thanks Bettina. Yes. De growth seems to be the ultimate aim of all this. It certainly seems the likely consequence.
I’m just not sure where to take all this tbh. I’m spending a lot of time on Substack as I get less busy elsewhere. But it doesn’t really offer much of a (financial) return in terms of time invested.
Merry Christmas to you Bettina. And thank you so much for your continued support. It is genuinely very much appreciated.
AI got the image a bit wrong - TwoTier is a lot chunkier than that.
Funny how TPTB in the UK want to take us back to a pre-science medieval life, without the religion (except Woke) and without electricity and - oh yes - real feudalism. Back in your box peasant!
Merry Christmas to you, LSO, and Happy New Year - I hope the 'crossroads' doesn't take your Substack away!
I wholly agree with you that "if it saves just one life" is about the worst reason possible for banning anything. We would have no food, shelter, power or transport.
Normally the issue with technology is overspecification. The camera on my phone would have beaten most of the kit David Bailey used in his heyday, but the photos don't. Partly because they are of things like hard to reach consumer units so I can read the small label telling me which MCB turns what things off. I don't do cat videos.
The reverse is true with AI at present, because it isn't intelligence at all. It doesn't have the capacity to understand why the question is asked, and so it just regurgitates stuff from the internet, and makes up things to fill the gaps. You can't beat it for telling you how to write a complex formula in Excel, as long as you know how to ask the question, but ask it to write a short story with Poirot as the main character in the style of Agatha Christie, and the result is pathetic. This is probably why those studying the impact of feminism on glaciology at Nopoint University get AI to write their garbage, and their "professors" get AI to mark the AI.
Frankly, AI today is not much more than a slightly better search engine for specific things, or yet another endless time waster like computer games and cat videos.
Of course, it may one day become intelligent, at which time we will have to dust off those old Terminator DVDs to work out how to deal with it.
Absolutely Jeremy. And yet they always use it as justification for the most draconian restrictions. I heard the ‘just one life’ excuse endlessly during lockdowns. It was incredibly disingenuous.
The idea that a LLM could become sentient is equally bonkers. Sure they can be impressive as you say, though only if you plug in the ‘correct’ inputs. The idea that they somehow ‘understand’ what any of the terms are is nonsense. They are just a’ Chinese Room’ if you know that idea.
I have to say though, I love video games and cat videos. Two of my very favourite time wasters.😆
ATB Jeremy and have a great Christmas.
And to you LSO.
Nice one. Like, could I run a space program from my laptop? Hmm...
Not often I have reservations about the conclusions and observations that you draw, LSO, but I could not be more cautious and pensive about the inexorable march of AI.
Here we are, humans, possessed of arguably the finest intelligence in history (if we set to one side that of dolphins and orangutans, both perfectly in balance with their surroundings and essentially kindly in their nature) and we are seemingly content to let the thumb-enriched players of computers to dictate how matters are to be developed in the future.
AI cannot scent the rain, nor see the love in a dog's eyes ... I tend to stick the to the real and the stuff not dependent on the click of a computer - but then I am very old and rather old-fashioned, and therefore not entirely to be trusted on matters in this new age.
I’m delighted to have this discussion with you no-one. I have a huge respect for your position. It is shared by many people. Not least Elon Musk. Who seems to know a thing or two about this stuff.
However I worry that the threat of AI, at least at its current level, is being massively talked up by governments as an excuse to censor and monitor adults. It is the perfect labour saving tool of authoritarians.
But too dangerous to allow into the hands of little people.
I can see ‘Have you not seen Terminator?’ Becoming the new ‘Won’t you think of the children?.’
But certainly this is a discussion we should be having. And thanks for having it with me.
Indeed, LSO. Digital impulses may be capable of many things but when I read of the electronic, blisteringly fast cogitations of AI, and the essentially plagiaristic nature of its pronouncements, I wonder if it could ever truly understand the humanity of someone like Alexander Pope:
"All nature is but art, unknown to thee;
All chance, direction, which thou canst not see;
All discord, harmony, not understood;
All partial evil, universal good:
And, spite of pride, in erring reason's spite,
One truth is clear, Whatever is, is right."
I’d never heard of the Carrington Event before. I’m going to look into that some more. Fascinating.
My dad could put a gearbox in a car and plumb and tile a bathroom from scratch. I have barely any of those skills yet I’m practically an engineer compared to my peer group.
The skills you have been pretty much eradicated from our culture. I believe for practical purposes we are all cyborgs now. Only capable of sustaining ourselves with the aid of machines.
Another ‘event’ and I’d only survive until my dried fruit and nuts run out. About a week if I was careful and didn’t share any with my kids …🤣
With regard to the plagiaristic nature of AI art I think Pope would immediately see the danger of all this then.
That currently AI is feeding/learning from human created art. But it will reach a point when most of the ‘art’ online will be AI generated. At that point it will become an ouroboros, forever eating a regurgitating its own AI tail. Weird. What will become of us then?
I’m intrigued as I am worried to be honest. But maybe I’m just being foolish.
Never foolish, LSO. All electronic 'stuff' is vulnerable eventually, if not to the simple 'off-switch' then to a Carrington Event (last witnessed in 1859 and long overdue a repeat, apparently. Ref: https://en.wikipedia.org/wiki/Carrington_Event)
For myself, having always been a bit of an an outdoors chap, so to speak, I have made sure that my children and grandchildren can forage for food, light a fire in the damp early morning, find water where there appears to be none, climb a cliff without ropes, ride a horse, and put an arrow into the gold at fifty yards. Call it neanderthal primitivism if you will, but I suspect it will serve them better in the near to medium term than any amount of button-pressing as we venture into the coming collectivist nirvana.
AI is an interesting one: for folk like me it generates 90% of my code and then I fill in the complicated bits - by asking it more questions. It also explains what it's doing and I comment throughout and look really clever, which I'm not.
AI for government and statist organisations is an absolute terror, so no wonder the EU and Left have banned it. Imagine a genuine AI system operating solely from logic, collating thousands of data points per millions of folk in seconds (doing away with the department that is suppose to do it) and simply saying 'here's your parking ticket at a football game. Here's you passing through the turnstile. Here's your gym membership and you go to this class regularly. Your welfare payments will be stopped'.
Or, the real terror for big government: "this department does not need 20,000,000 quid. Forty percent of those here do nothing of value. The department budget will be reduced by 70% as the 50 quangos and charities who demand money of it do not need to exist either."
Imagine AI's opinion of the EU! A pointless talking shop that has hindered economic growth, prosperity and freedom for no benefit but that of those inside the thing.
We're a long, long way from true, aware thinking AI but considering where we are and where we will be and government should be afraid. As it is we are witnessing the last flailing abuses of a system so self serving, arrogant and utterly terrified of losing the control it is frantically giving what little relevance it has away to a body utterly untouchable by the public.
Great points Wibbling. I’m pretty sure that the bosses are aware of them all. Which is why they won’t take their eye off the ball like they did in the early days of the internet, allowing it to become, for a while, a proper free for all, which released all the innovation and groundbreaking services we see today.
Services like Google maps. YouTube. Paypal.plus social media sites like Facebook would have been throttled at birth if the regulators had been as in charge then, as they are now.
The idea is clearly to take charge of AI at this current level. And use it to police our online communications and limit our interactions. Flagging ‘problematic’ content and deleting any form of dissent it before it can infect the minds of free thinkers and sceptics.
Big fat state will try. It'll try legislation, oppression, suppression and when all that fails outright control; but it *will* fail.
Somewhere, someone will ignore their stupid dictat, maybe in another country and that'll be it. Government is like the courtiers for Canute, although rather than believing the king can hold back the tide, they think they can.
The state machine is wrong. Massive job losses, the WFA, high unemployment, inflation, debt, rich folk leaving, industry evaporating, expensive food, no growth (and the inevitable recession) are all results of sky high energy prices. Government is a cartoon character jamming fingers in the pipe it's drilling through and failing utterly to stop the problem they're causing. These people are dangerous, stupid, mendacious, spiteful, egotistical, greedy, incompetent, vicious, deceitful and arrogant. That is why everything they do fails. AI will be no different.
I completely agree Wibbling. Although it doesn’t seem to be failing at the moment. Unless you mean it’s failing us!
It seems clear to me that the biggest issue facing any independent-minded individuals today is the provenance of digital evidence. When AI can increasingly generate images and videos that can not easily be detected as generative, we know longer can know much of anything. I'm envisioning a world in which someone can, say, come up with a video clip that purports to be from the bodycam of a police officer doing and saying virtually anything the prompter wants. If we don't solve this problem, governments are going to use it as a way to control or shut down all the foundations of our independence.
To some extent this is already happening Jeff. Maybe not the creation of images but certainly the manipulation of data.
As you can probably see I work hard to link supporting or salient articles to my pieces. Google undoubtedly uses AI to downvote or ‘hide’ the ‘wrong’ sources.
If I look for anything vaguely controversial or a ‘conspiracy theory’ say Climate Change for example. Then a search for dissenting voices is usually fruitless unless you already have a source, like Spiked! for instance.
I have a lot of time for different opinions on this subject. But it is increasingly clear to me that the threat isn’t so much ‘AI’. It is ‘AI plus government’.
Google specifically excludes and 'deranks' (if such a word exists) sceptical arguments. For those of us wanting information the relentless promotion of political opinion merely demonstrates how little there is.
It definitely seems to from my perspective. I try and save as much as I can in terms of links when I first see a story. Because if I don’t at the time, I know it will prove v difficult to find them later.
I stopped using Google for search and everything else when it became clear they were manipulating it. It can be challenging. I use Presearch for search. It rarely disappoints. It's an open-source disributed search engine which funds server costs with minimal advertising. They've got some kind of crypto currancy reeard sysrem for viewing ads which I ignore. Try presearch.
Duckduckgo apparently doesnt do much native search themselves, but runs your search through Google but strips your identifiers so the latter can't track you. You still get Google's bias. Google likes DDG because it protects them from monopoly attack.
We've reached a point as a nation where double standards are rife and obvious, with no punshments for those abuses of justice and morality. In this world, truth is no longer a defence. The state doesn't need to fabricate evidence. Punishment depends on affiliation (for or against big government), not for actual crime.
Very interesting Jeff. I’ve always found DDG search next to useless. I’ll try pre search. Thanks!
I use AI generated stuff at work and refer to it as Artificial Stupidity it's that rubbish. It's only as good as it's programmed to be, garbage in etc. Totally agree re Poundshop Voldemort. He's a traitor to this country and should be dealt with accordingly. I can't express in words how much I despise him. I think there's a reason he's rarely in the UK since he took office a) he hates the UK b) he knows by now how much the population loathes him. We're all getting recession for Christmas and he may have concerns about what the fallout will be....
Ha ha. At its current level, AI is certainly not the existential threat to our society that Starmer and his ilk represents. Merry Christmas Toffeepud.
AI is doing great things with hearing and visual aids for those impaired in those departments.
Or perhaps I should say, "enabling great things".
That’s really interesting Brendan. I hadn’t really considered that before, but of course, it’s a perfect fit. More stuff like this would be great. There’s so much potential, along with the inevitable catastrophising.
I happened to be at last years "British Machine Vision" conference (playing in the dance band after dinner) and got to watch some of the presentations and speak to some of the attendees, from all over, not just Britain.
"AI" is a broad generalisation for a collection of developments including neural networks etc. It was surprising and interesting to me as it probably is to you. That's an indication of how poorly the news media is serving us on this topic. Which is also a function of how technically illiterate they are.
Look up BMVC2023.org and you'll find the conference proceedings online for this and other years. Just scanning through the synopses of the papers will give you an idea of the breadth and depth being covered.
Not completely with you on this. US corporations have prioritised their profits over children’s lives when they have refused to resource changes to their social media algorithms which pull children to harmful and self-harm material. National governments need to bring home their responsibility to them.
Very fair point Jaffa. But governments won’t use these powers to protect children. Despite their protestations that this is their only motivation. Instead they will use these powers to censor, and to remove the rights and freedoms of adults.
I’d rather at least those points of the Act were there, so they can be used to prosecute corporate killers
We must embrace AI. If the sane among us shun it, the feral left will gain even more ground. Like a lot of things, it can be used for good and bad. It's a shame we don't have Sora. I am getting a bit fed up with using Bing AI for my work. They make some really good pictures, but too often there is a incongruous or bizarre thing thrown in that ruins the image.
I tend to agree Adam. We are at a junction in terms of technology. The best way to stop the government using AI against us is probably (though I’m not nailing my colours to the mast in this) making it available for everyone.
It’s a bit like the argument that Americans Second Amendment protects the First. Not sure I’m entirely convinced but even so it is a compelling argument.
I expect the US government will at some point in the next two years attempt to class some forms of advanced AI as ‘munitions’. That way it can set all the parameters for AI use that it wants.
This was exactly the argument David Brin made in his 2000-something book, "The Transparent Society". He went into the research for that book looking for reasons why we need strong guarantees of privacy. What he discovered while doing the research completely turned his thinking around. His conclusion was that if we try to regulate things like privacy (as the EU is currently doing) all that will happen is that governments and the rich and powerful will continue to use the technology that violates it, while the rest of us will be forbidden from being able to also use it to fight back. My paraphrase of his argument is grossly over-simplified. I recommend the book, even though it is old now.
Thanks Jeff. It sounds highly prescient.
We keep handing government more and more powers to ‘keep us safe’. But we end up no safer than we were at the beginning. We just have less power to do anything about it.
We don't exactly "hand" government more powers, they just keep taking them... whilst enough keep voting for the sort of government that will ensure the practice continues, at least in the UK. OTOH, they don't even provide the safety that most would value, such as protection from crime.
A Very Happy Xmas LSO and a more prosperous 2025!!!
Absolutely Alexei. And they keep ‘devolving’ power away from people we can vote for, to people no one can vote for. And calling it ‘Our Democracy. ‘ 🤣
Happy New Year to you and yours Alexei. And thanks so much for coming! 👍
The Quangocracy at full force. My opinion as it is, is that if body such as this oversteps its original mandate but is still considered necessary, the public gets to vote on who runs it and how much funding it gets. Directly. Because it has become a de-facto arm of the state.
We can also vote to disband it.
Most if not all of the functions can be handled simply by the government responding to citizens suggestions or complaints.
It’s a good plan Overhead. My worry with all this stuff is how easily it gets captured. Just thinking this morning about all the ‘independent’ institutions which aren’t. And how ‘devolving’ democracy to ‘the local level’ often leads to less democracy.
I think ultimately we must pare back the state significantly. Give it much much less to do, and relentlessly hold the people doing those things to account.
But yes. I’d take your plan over the current stitch up. ATB
Hi LSO, good piece. Generative AI is in its infancy. There’s going to be lots of dead ends, but when the dust settles, just like the dot.com era, it will change the way we work (already is in some sectors). That of course, is if there is much of an economy or for that matter country left after 5 years Labour to cap off 14 years of the Tories. As the Chinese curse would say, we live in interesting times.
Wishing you a Merry Christmas
Hi P. Yes there’s no putting the genie back in the bottle. for better or for worse. Who knows where this stuff will lead us? All I know is, if we let the government convince us it’s too dangerous for us, but OK for them, then we’ll end up nowhere good.
Merry Christmas to you and yours, and thanks for all your support in 2024. Much appreciated.
Nobody's going to build an AI data centre in England when electricity costs 4x what it costs in the US.