AI to make life richer and more meaningful

Page 1 of 2 [ 17 posts ]  Go to page 1, 2  Next

kokopelli
Veteran
Veteran

User avatar

Joined: 27 Nov 2017
Gender: Male
Posts: 3,634
Location: amid the sunlight and the dust and the wind

18 Mar 2023, 3:26 pm

We already have things like Alexa that can answer limited questions. If you have an Alexa device, say to it, "Alexa, you sure are a lot of fun."

With much more AI, imagine how we could potentially have computers that can take on multiple personalities and engage in meaningful discussions with us. For example, suppose you like baseball and your favorite team is the Arizona Cardinals. Then the AI could potentially generate multiple personalities with different viewpoints to discuss the Arizona Cardinals. It could even tie multiple conversations together so that the viewpoints express in your discussions with the AI personalities could incorporate viewpoints expressed by others in their discussions. It could also give a running description of the current game if you can't go the game or watch it on tv.

Same thing with other areas. At a recent event, I had pork chops a couple of weeks ago that were imaginably good and I have no idea how to make them. I see the chef once a year at that event. If I remember, I'll ask him about it next year. But what about a discussion about cooking with AI generated personalities that could give cooking tips and techniques to make those same pork chops.

Take it even further into prisons. I've never been in jail or prison but I'm sure that the lack of things to do would drive me crazy. But what if each cell had a speaker and microphone safely contained within the wall that would allow the inmates to discuss issues with AI generated personalities. Of course, the AI personalities would have some things put off limits -- for example, someone in prison for torture and murder should probably have limited access to discussions about those.

For most prisoners, though, there could be AI personalities to discuss just about whatever they wish as well as to "read" today's news to them or discuss news of the past. The AI could also act as an intermediary between the prisoner and his family members by passing on information each way. For example, a prisoner could be informed by the AI within minutes or hours that his sister just had a baby and various things about the baby such as the name, birth weight, and whether nor not there were triplets.

In either case, prisoners or free citizens, create an environment designed to allow each to enjoy life more. They might not even be able to tell when they are talking to a real person and an AI person. To work right, it would need to be a whole roomful of AI personalities available, not just one Alexa.



Cornflake
Administrator
Administrator

User avatar

Joined: 30 Oct 2010
Gender: Male
Posts: 65,512
Location: Over there

19 Mar 2023, 8:55 am

kokopelli wrote:
We already have things like Alexa that can answer limited questions. If you have an Alexa device, say to it, "Alexa, you sure are a lot of fun."
Repeating it produced: "Sorry, I don't know that one" or "Sorry, I'm not sure" or "Sorry, I don't know that".
Did I win? :scratch:

You could once send Alexa into a nosedive with "Alexa, what's the factorial of 100" and in response the entire numerical result was recited, one digit at a time. This went on for quite a while...
Saying "Alexa, stop", sometimes repeatedly, terminated the recital.

Now the response is "9.333 times 10 to the 157th power".
Pah. Nowhere near as amusing.

Generally Alexa is a bit dim and will often say "Here's something I found on the web" - and then read out what I could have found for myself if I wasn't being lazy.
It's reasonably useful here as a voice-operated switch turning lights/kettle etc. on or off or, rarely, reading the news headlines (from one of only three sources) or playing music.

At some point the ability to remember the current state of a switch was added.
"Alexa, is the kettle on" would give "Hmm, I don't know that one" but now is able to say "The kettle is off" - hardly a giant leap for mankind, but it's in the right direction.

It's supposed to be able to control my smart TV but that appears to be a work in progress.
The response is often "Hmm, something went wrong" or "That service isn't available" - usually followed by whatever I'd requested working anyway.
Thanks - I'll stick to the remote control.

I think there's a way to go before AI, at least in this form, can enrich life and make it more meaningful - but the progress is nevertheless interesting.


_________________
Giraffe: a ruminant with a view.


klanka
Veteran
Veteran

Joined: 31 Mar 2022
Age: 46
Gender: Male
Posts: 1,888
Location: Cardiff, Wales

19 Mar 2023, 9:46 am

It'll drive more people off the internet.

The dating and other websites have been taken over by scammers

Once the ai is more widespread you'll either be talking to ai or a scammer whenever you try to connect with someone online.
So maybe a few more meetings in real life will take place.



kokopelli
Veteran
Veteran

User avatar

Joined: 27 Nov 2017
Gender: Male
Posts: 3,634
Location: amid the sunlight and the dust and the wind

27 Mar 2023, 3:58 am

Cornflake wrote:
kokopelli wrote:
We already have things like Alexa that can answer limited questions. If you have an Alexa device, say to it, "Alexa, you sure are a lot of fun."
Repeating it produced: "Sorry, I don't know that one" or "Sorry, I'm not sure" or "Sorry, I don't know that".


That's interesting. The local preacher has one. It let out a long "Wheeeeeeeeeeeeeeeeeee" or something like that when I said that to it.

Quote:
Did I win? :scratch:

You could once send Alexa into a nosedive with "Alexa, what's the factorial of 100" and in response the entire numerical result was recited, one digit at a time. This went on for quite a while...
Saying "Alexa, stop", sometimes repeatedly, terminated the recital.

Now the response is "9.333 times 10 to the 157th power".
Pah. Nowhere near as amusing.


I wonder what it does if you ask for the square root of four to a million places.



Cornflake
Administrator
Administrator

User avatar

Joined: 30 Oct 2010
Gender: Male
Posts: 65,512
Location: Over there

27 Mar 2023, 6:54 am

kokopelli wrote:
Cornflake wrote:
kokopelli wrote:
We already have things like Alexa that can answer limited questions. If you have an Alexa device, say to it, "Alexa, you sure are a lot of fun."
Repeating it produced: "Sorry, I don't know that one" or "Sorry, I'm not sure" or "Sorry, I don't know that".

That's interesting. The local preacher has one. It let out a long "Wheeeeeeeeeeeeeeeeeee" or something like that when I said that to it.
There might be variations to the responses, localised by country or maybe even state, in an attempt at sounding more human.

Quote:
I wonder what it does if you ask for the square root of four to a million places.
Nah - looks like the Alexa droids have cut off the amusement with numbers. They're onto us! :lol:

Asking that just gets: "A quadrillion is followed by quintillion on the numeric scale", so it didn't answer the question at all.
Asking for pi to a million places gets "An approximation of pi is" then a recital up to 22 places, followed by "Alrighty! I've gotten you this far. I'll let you take it from this point".


_________________
Giraffe: a ruminant with a view.


Tim_Tex
Veteran
Veteran

User avatar

Joined: 2 Jul 2004
Age: 44
Gender: Male
Posts: 45,472
Location: Houston, Texas

27 Mar 2023, 11:08 am

While I think AI can rectify many issues we face, I also have concerns about it replacing human workers on a massive scale. I read that ChatGTP could put half the global work force out of a job.

The bottom line: AI, automation and ChatGTP can work, but only when done responsibly and ethically, and after considering human needs.


_________________
Who’s better at math than a robot? They’re made of math!

Now proficient in ChatGPT!


Cornflake
Administrator
Administrator

User avatar

Joined: 30 Oct 2010
Gender: Male
Posts: 65,512
Location: Over there

27 Mar 2023, 11:19 am

Tim_Tex wrote:
I read that ChatGTP could put half the global work force out of a job.
That's extremely unlikely - ChatGPT is still only a bunch of canned responses.
It's a rather large bunch though, and that's what helps make it look pseudo-sentient.


_________________
Giraffe: a ruminant with a view.


Jono
Veteran
Veteran

User avatar

Joined: 10 Jul 2008
Age: 43
Gender: Male
Posts: 5,603
Location: Johannesburg, South Africa

27 Mar 2023, 11:50 am

Cornflake wrote:
Tim_Tex wrote:
I read that ChatGTP could put half the global work force out of a job.
That's extremely unlikely - ChatGPT is still only a bunch of canned responses.
It's a rather large bunch though, and that's what helps make it look pseudo-sentient.


I don't think it's canned responses. Responses are generated based on a model.



Cornflake
Administrator
Administrator

User avatar

Joined: 30 Oct 2010
Gender: Male
Posts: 65,512
Location: Over there

27 Mar 2023, 3:25 pm

My point is that there's no original thought, despite appearances.


_________________
Giraffe: a ruminant with a view.


NeilM
Toucan
Toucan

User avatar

Joined: 20 Aug 2016
Age: 72
Gender: Non-binary
Posts: 277
Location: Virginia, USA

28 Mar 2023, 12:44 pm

Everyone--meet Moxie!

https://embodied.com/

I think this is more what the op had in mind. It is designed to help the child who cannot make friends, help them get so then can make some friends. Moxie isn't new; its been around for several years.


_________________
Your neurodiverse (Aspie) score: 120 of 200
Your neurotypical (non-autistic) score: 74 of 200
You are very likely neurodiverse (Aspie)

AQ = 38 MBTI = ISTJ Gender = Non-binary
I strive not to perseverate. You can PM me for more info.


Lost_dragon
Veteran
Veteran

User avatar

Joined: 6 May 2017
Gender: Female
Posts: 2,738
Location: England

28 Mar 2023, 5:35 pm

It concerns me. The direction that we're headed. Sadly I don't think there's much stopping it. I think it's been a long time coming.


_________________
24. Possibly B.A.P.


RandoNLD
Toucan
Toucan

Joined: 16 Mar 2023
Age: 35
Gender: Male
Posts: 291
Location: 90º north Lat, 90º south Late

01 Apr 2023, 9:02 am

...or amplify the dystopia we live in.



Double Retired
Veteran
Veteran

User avatar

Joined: 31 Jul 2020
Age: 69
Gender: Male
Posts: 5,141
Location: U.S.A.         (Mid-Atlantic)

02 Apr 2023, 10:37 am

Lost_dragon wrote:
It concerns me. The direction that we're headed. Sadly I don't think there's much stopping it. I think it's been a long time coming.
I worked in computer security and that aspect of this bothers me.

I think it is wonderful that our stuff will get smarter. But I don't like all of the interconnections. I see no reason for my desktop computer, my bride's phone, our TV, our vacuum cleaner, our thermostat, and my car all to be interconnected and connected to the Internet.

I would like it to be autonomous intelligence.

And I would set expectations reasonably. I think it is great that my car is autonomously smart enough to understand and act appropriately when I tell it "Play album Koyaanisqatsi" or ask it "Where am I?" but I see no need for it to be able to answer "What is the capital of Columbia?"

Does Alexa know how to get to Wrong Planet and what star it orbits?


_________________
When diagnosed I bought champagne!
I finally knew why people were strange.


cyberdad
Veteran
Veteran

User avatar

Joined: 21 Feb 2011
Age: 56
Gender: Male
Posts: 34,146

02 Apr 2023, 3:57 pm

There's a number in the creative industries who aren't as enamoured by AI
https://www.iotforall.com/impact-of-art ... job-losses

Even in my field of education the prospect of AI bots leading online education which draws on millions of databases of source material is inevitable. As of 2023 we already know students are using AI (ChatGBT) to write essays and assignments.

Some disciplines in STEM that require laboratory work will persist longer but they will eventually be replaced by a combination of AI and automated robots (Watch Japan, they will be first).



RoadRatt
Veteran
Veteran

User avatar

Joined: 26 Aug 2014
Age: 59
Gender: Male
Posts: 54,197
Location: Oregon

02 Apr 2023, 5:52 pm

I have zero interest in ever communicating with AI. I would never be able to trust it.


_________________
No power in the 'verse can stop me. - River Tam (Firefly)


Lecia_Wynter
Velociraptor
Velociraptor

User avatar

Joined: 16 Dec 2022
Age: 34
Gender: Female
Posts: 411

30 Apr 2023, 9:10 am

kokopelli wrote:
We already have things like Alexa that can answer limited questions. If you have an Alexa device, say to it, "Alexa, you sure are a lot of fun."

With much more AI, imagine how we could potentially have computers that can take on multiple personalities and engage in meaningful discussions with us. For example, suppose you like baseball and your favorite team is the Arizona Cardinals. Then the AI could potentially generate multiple personalities with different viewpoints to discuss the Arizona Cardinals. It could even tie multiple conversations together so that the viewpoints express in your discussions with the AI personalities could incorporate viewpoints expressed by others in their discussions. It could also give a running description of the current game if you can't go the game or watch it on tv.

Same thing with other areas. At a recent event, I had pork chops a couple of weeks ago that were imaginably good and I have no idea how to make them. I see the chef once a year at that event. If I remember, I'll ask him about it next year. But what about a discussion about cooking with AI generated personalities that could give cooking tips and techniques to make those same pork chops.

Take it even further into prisons. I've never been in jail or prison but I'm sure that the lack of things to do would drive me crazy. But what if each cell had a speaker and microphone safely contained within the wall that would allow the inmates to discuss issues with AI generated personalities. Of course, the AI personalities would have some things put off limits -- for example, someone in prison for torture and murder should probably have limited access to discussions about those.

For most prisoners, though, there could be AI personalities to discuss just about whatever they wish as well as to "read" today's news to them or discuss news of the past. The AI could also act as an intermediary between the prisoner and his family members by passing on information each way. For example, a prisoner could be informed by the AI within minutes or hours that his sister just had a baby and various things about the baby such as the name, birth weight, and whether nor not there were triplets.

In either case, prisoners or free citizens, create an environment designed to allow each to enjoy life more. They might not even be able to tell when they are talking to a real person and an AI person. To work right, it would need to be a whole roomful of AI personalities available, not just one Alexa.


We do need prison reform but that upgrade you chose doesn't sound particularly utopian.

A problem with right-wingers is they are too hard on criminals while leftists let criminals do whatever they want and then you get cities like Portland.

The only rational scenario is to force feminize criminals since 90% of criminals are male. The idea right wingers have that "prisoners should suffer" isn't very Christian of them.