Today's discussion about the news
Comments
-
You can confirm things online with a simple message to a verified source.
0 -
It's on the internet so it must be true. Obs.
PS - I won the 1948 version of the Peace Race. This wasn't publicised by the East as it was bad news and the West didn't recognise it. True story as it's on the internet.
The above may be fact, or fiction, I may be serious, I may be jesting.
I am not sure. You have no chance.Veronese68 wrote:PB is the most sensible person on here.0 -
Not sure that's true. My memory is 30+ years ago most households bought the local newspaper and having worked for ours for a short period it was staffed with quite a few career journalists that would have been on top of stories like this. Nowadays our local paper has only a small local office - it's not even based in our city - and nobody I know buys it. Most of its output on the internet is non local stories designed to get clicks - going in depth on stuff like free ports or local politicians not telling the truth isn't going to do that.
[Castle Donington Ladies FC - going up in '22]1 -
No, that's why it won't work, because people make stuff up and it can't tell the difference.
0 -
Is this just Reuters with some AI buzzwords/catchphrases.
0 -
How do you tell the difference?
0 -
No
0 -
Answer, check multiple verified sources.
0 -
Who's verifying the source? Is the algorithm told which sources to trust? By whom? A whole lot of people were convinced the Southport attacker was Muslim and an asylum seeker because they trusted the source.
Would you trust the Musk, who's reposted numerous bits of false information, to set up 'trusted sources' for Grok?
1985 Mercian King of Mercia - work in progress (Hah! Who am I kidding?)
Pinnacle Monzonite
Part of the anti-growth coalition0 -
Another day, and another day when then Telegraph morphs even further into a satire of itself. A newspaper in name only.
0 -
Well you're constantly quoting twitter so clearly you do. Do you hop on a plane visit the source and make sure they're telling the truth?
0 -
If it's one source it simply rejects the information. All this can be evolving making it more accurate.
The computer you're using isn't the same one you started with neither is the software. It's quicker, more features, more reliable, more efficient, probably cheaper.
I don't understand why you look at today and think it's the future. That's not reflected in history, especially when it comes to technology.
0 -
If AI replaces journalists who will be these verified sources it will use? It can also hoover up alternative views to avoid bias but that doesn’t mean both views aren’t incorrect.
0 -
Anybody online who is their persona in real life, let's start there.
0 -
Twitter isn't a source: it's a platform. I don't quote Twitter; I quote something that someone has posted on Twitter. Generally, I have some degree of trust that what they have written is reasonably accurate, but it's just trust, not verification. I'm aware that it may not be accurate and that both I and the original author have our particular biases. If I did want to verify something that requires work. Maybe not getting on a plane but it does require collecting information from a variety of primary sources and making value judgements on those pieces of information as to which is a more accurate reflection of reality.
Current LLMs have no concept of reality whatsoever and so produce claims that strawberry has two RS and people have random numbers of teeth and fingers. Until someone works out how to encode that 'reality check' that we all do hundreds of times a day without realising (and which still often lets us down when reality conflicts with our assumptions) there is no way AI can verify anything.
1985 Mercian King of Mercia - work in progress (Hah! Who am I kidding?)
Pinnacle Monzonite
Part of the anti-growth coalition0 -
Can't happen. AI cannot go out and find stories, all it does is regurgitate existing stories. AI needs a source, ie journalists.
The above may be fact, or fiction, I may be serious, I may be jesting.
I am not sure. You have no chance.Veronese68 wrote:PB is the most sensible person on here.0 -
How can those be verified? Bear in mind that plenty of online profiles are not real people, but pretend to be.
1985 Mercian King of Mercia - work in progress (Hah! Who am I kidding?)
Pinnacle Monzonite
Part of the anti-growth coalition0 -
Inc e.g. that f-zing-thing, which is actually a test bot from the increasingly murky world of the Muskrat.
Which you have now found on t'interweb so it must be TROOO...
0 -
I'm just thinking about the concept. Pretty much all of us quote from online content, it just seems the next logical step. The salient advantage being speed of being published and also being corrected if untrue.
Journalism as well knew it is just too slow, also, it takes on their perspective, how can it not.
0 -
You're missing out a key part of the process. We're not reposting stuff at random. We're making judgements about the veracity of those pieces of information when we choose to repost them. AI as it currently stands has no understanding of what veracity is, let alone being able to assess whether something is true.
1985 Mercian King of Mercia - work in progress (Hah! Who am I kidding?)
Pinnacle Monzonite
Part of the anti-growth coalition0 -
The scary thing is that the tech companies think like focuszing.
0 -
OK, how long until you think this could happen given the internet/commercial computers have been around for less than fifty years?
0 -
I just think it's possible given the rapid evolution of IT. In reality I'm still seeing potholes.
0 -
Given we still don't really understand consciousness, or emotion, I think AI might take a while to catch up with a lot of the filtering/understanding we do without consciously thinking about it.
It's not an unserious question I ask whether AI is capable of asking itself if it's being stupid. Also, I suppose, can it have hunches?
0 -
You'd have to say it comes to time as you know the rapid evolution is there.
0 -
Read up on the debates about consciousness: at the same time as we learn about the smallest and largest things in the universe with a fair degree of certainty, despite hundreds of years and some of the greatest minds trying to solve the riddle of consciousness, we're still nowhere near being able to explain it in anything other than colourful analogy.
As I've said before, there's no doubt that AI is going to make great strides in many ways in fairly short order, but at the same time it's going to be equally important to understand its limitations. I'd probably apply the same criterion to AI that I apply to people: I don't trust those people who don't question and doubt their own understanding: I'm always more likely to trust someone who is prepared to admit that they might be wrong, and at the moment I can't see how AI can do that.
0 -
I simply can't get anywhere near wrapping my head around how this can be a thing in the 21st century.... well, at any time, TBH.
0 -
Sure this will have been shared before. Afghanistan in the pre-Taliban 60s. Sometimes going forward in time is not progression. 😢
The above may be fact, or fiction, I may be serious, I may be jesting.
I am not sure. You have no chance.Veronese68 wrote:PB is the most sensible person on here.0 -
Just as a thought experiment, imagine if that edict was enforced on a type of animal... "Horses must be covered totally in black, even their hooves." Yeah, utterly, totally mad. But women? Humans?? What is it that the Taliban are so afraid of?
As I say, I simply can't even start to understand it.
0 -
Who wants to tell the Telegraph that the UK is still part of Europe?
0
