That’s what I meant when I said we should be celebrating but unless things change fairly massively then how will everyone get to share in this utopian dream? We’re going to have to start paying a decent wage for doing almost nothing. I don’t think there’s many people ready to get their heads around that.
Just realised, you’re right. This is the old version. Didn’t notice that you need the plus as Load pointed out.
Yeah, even the plus doesn’t get you to GPT-4. It’s just 3 plus…
Read somewhere it’ll be September before 4 will open up for public use. But then, wasn’t AGI meant to take another 7 years to begin to appear as predicted in February.
5 weeks later, AI advanced enough already that it is already happening.
*Don’t quote me on any of this ‘fact’. Ask an AI to verify it, or something.
I asked chatgpt, the cheap version to
‘Write me a rap about the struggles of an affluent white man in Stoke on Trent’
Answer…
Verse 1:
Yo, I’m an affluent white man from Stoke on Trent
Living life in the fast lane, you could say I’m content
But people judge me for my privilege and my wealth
They assume my life’s easy, like I ain’t facing no health
Chorus:
But yo, life’s hard for me too, can’t you see
Just 'cause I’m white and rich, don’t mean I’m living free
I got my own struggles, my own battles to fight
So don’t judge me 'til you walk a mile in my Nikes
Verse 2:
Growing up in Stoke, I had it pretty good
I never had to worry 'bout food or neighborhood
But that don’t mean I ain’t got my own demons to face
Like anxiety and depression, trying to find my own place
Chorus:
But yo, life’s hard for me too, can’t you see
Just 'cause I’m white and rich, don’t mean I’m living free
I got my own struggles, my own battles to fight
So don’t judge me 'til you walk a mile in my Nikes
Verse 3:
I see people judging me, thinking I got it all
But they don’t know the pressure, the expectations that fall
On my shoulders, from my family and my peers
Trying to keep up appearances, hiding all my fears
Chorus:
But yo, life’s hard for me too, can’t you see
Just 'cause
It got stuck here and stopped working haha
ChatGPT drops mic
I work on messaging platforms for a marketplace.
I built a thing recently that detects the language people are messaging in and will auto translate it to whatever other language the other parties are reading it in. They can then reply in their language and it will translate back into the original language. So one person could be typing French, the other German and they would be able to type back and forth in real time. All whilst our support and admin staff can jump in and view it in English. All super easy with the new services that are cropping up. Just smartly talking to a number of APIs and hooking it up.
Wife already uses a chatGPT subscription so a developer working for her can use it. He created some kind of Google Fit integration with their project and it took him a day to complete with ChatGPT rather than weeks.
We talked about the implications of AI last night, she’d thought about it in much different terms than I had, and we discussed:
- Obviously, real/fake - it’s like going back to the stone age, we have to see/hear what a person has to say in person to know it was really said
- Porn (obvs) - apparently people are already attempting to deepfake children’s faces onto existing child porn (videos)
- Diseases - creation of diseases to target specific races/bloodlines
- War/weapons - that are absolutely clinical in targeting specific people
If you’ve had any exposure or are looking into AI, all the above are on the table right now. It’s absolutely terrifying. We are at an absolutely pivotal moment in history IMO
but the reverse of all of those is also true. use AI bots to crawl the web for illegal porn, use AI to find cures for viruses/diseases, use AI for defence systems etc.
if it’s the god machine that people are trying to tout it to be (no doubt to drive up share prices) then it should be able to cure everything, solve world hunger and create world peace
By all means we should already there. At this stage scarcity and the concept of having to work 40 hours a week to exist are artificial.
Could have had that years ago but wheres the money in it? - big pharma and military industrial complex.
Pharma will be doing everything they can to use A.I and patten the shit out of what their A.I solves/cures
I work as a translator. Fortunately for legal documents in particular (which are often used in court) an experienced linguist is required to translate them from scratch. The problem with AI is it doesn’t understand the subject specific context of certain words and therefore the meaning and stye of the document is lost in translation. I don’t envisage a future where human translators will be completely redundant but I may be wrong which is worrying. Certainly for basic company promotional material you could just use google translate and then just lightly edit the result to make it more natural and readable.
As an experiment I’ve compared my translation from scratch of an excerpt of a legal contract to google translate’s version:
15 – RESILIATION
Dans le cas de manquement de l’une des Parties à l’autre quelconque des obligations résultant du Contrat, l’autre Partie pourra la mettre en demeure, par lettre recommandée avec accusé de réception, et demander de réparer ce manquement dans un délai maximum de trente (30) jours calendaires.
My translation:
15- TERMINATION
In the event of failure by one of the parties to perform any of the obligations under the contract, the other Party may give that Party formal notice, by registered letter with acknowledgement of receipt and require that the breach be remedied within a maximum of thirty (30) calendar days.
Google translate:
15 – TERMINATION
In the event of breach by one of the Parties of the other of any of the obligations resulting from the Contract, the other Party may put it on notice, by registered letter with acknowledgment of receipt, and request that this breach be repaired within a maximum period of time. thirty (30) calendar days.
You can hopefully see Google translate is at this point far worse than a competent translator. But I wonder how quickly the gap will be breached. For a basic letter or tourist material the gap wouldn’t be as wide.
This is a translation by chatGPT:
“15 - TERMINATION
In the event of a breach by one Party of any of the obligations arising from the Contract, the other Party may give notice to the breaching Party, by registered letter with acknowledgement of receipt, and request that the breach be remedied within a maximum period of thirty (30) calendar days.”
That’s worryingly good. It’s lost none of the meaning of the original and the style is pretty good too. It’s better than both Google translate and mymemory translated, which is another service I tried. If that’s the general quality of chat gpt translation then I can totally see why a lot of companies are using it instead of human translators. I wasn’t expecting that.
I think that’s chatGPT 3.5 as well
We say things like ‘translate this in an informal tone’. Google can’t do that.