Good onya 
. I guess itās part of your human being āsurvival skillsā 
Escape the crap before it becomes āinsaneā. 
I eventually have the same sort of āphysiologicalā reaction 
And the tutorial looks fun. Much more fun then discussing with the fancy AI, if you ask me. Enjoy the moment and have a great sunday 
1 Like
Hello @Deltakosh, since chatGPT gives more and more better answers around questions about babylon.js in this forum the idea is to create an auto-answer plugin, which everyone can turn on or off to get immediate answers to his/her questions, to avoid long waiting times. What do you think about?
The software of this forum seems to be open source: Is Babylonjs Forum opensource? - #3 by RaananW
We actually thought about it but we did not implement it for several reasons:
- the forum is hosted by Discourse and we cannot get the plugins nor the update we want as they maintain a SLA for us to ensure that the forum is always available
- we have no budget to support the API cost of ChatGPT
- we would have to add a clear disclaimer as the answers will not be vetted by us. That would also be a source of confusion for users
5 Likes
As an addition,
ChatGPT is great,
but often the answers, atleast when related to code, are flawed and wonāt work as intended without manual review and edits,
it still have a long road ahead.
so having something like ChatGPT respond to people who are trying to understand a piece of code or how to do something in babylon, and get non-functional or non-complete responses, will just cause them to become even more confused.
1 Like
This is a great example How to get rid of aliasing effect of clipping - #4 by sebavan
as it hallucinates some APIs and chose the wrong technique so human review should currently be mandatory before answering as @Deltakosh mentioned.
1 Like
And all the chatgpt answers from today were wrong or containing half of the answer being wrong so I am all for more validation before answering with it even if I love how structured and english the answers are compared to the best of my writing abilities 
1 Like
From now on, I will check the answers of the deus ex machina before I spread misinformation that leads to more confusion and frustration than intendedā¦ I apologize for any inconvenience.
1 Like
nah nw it is pretty good to try it and if you are unsure just run it in this thread and I ll check if all good ?
1 Like
I believe you understand that the average delay 
to get an answer in this forum is WAY BELOW any other. And then, the quality of the answers (including Yours) is also WAY ABOVE the average 
Personally, I do not need chatGPT (nor any other AI) in this forum. If I want chatGPT (and I sometimes do) then I have chatGPT already. The reason I come here is to get an answer from human people, users and experts of BJS. I see absolutely no interest at all in getting pre-formatted answers from a bot. Of course as always, my opinion only 
I can assure you youāre at least not the only exception. As a child, I also wanted to be an adult and mostly made friendship with much older persons and now that Iām supposed to become āa reasonableā old adult, I tend to want to become a child again And Iām quite sure there are others like us 
1 Like
Two ways to improve better answers in future:
-
you have to tell the AI that the code is wrong and how it would have been correct.
-
AND/OR you have to add to the core functions those functions that the AI claims exist.
1 Like
Yes, in essence yes. But the problem with the current version AI is that nobody (including the creators) is able to tell just what exactly makes the AI āchange opinionā or in other words ācreate a new patternā.
As I said before in this forum, the next challenge is all about āeducatingā and ānurturingā the AI. But for this to happen in a way that remains āunder controlā or āunder surveillanceā the core should be a lot more ātransparentā and āaccessibleā then it is today.
It would be interesting (and somewhat frightening) to see what it does. I mean, most of us, deep-inside, believe that our āmindā or āspiritā goes beyond our brain and our physiological human being. What if we would be able to prove that this is not the case?
And I was just replying to your daydreams 
I feel that even if science could prove it, there are just too many powers that want us to keep with our ābeliefsā 
The story has been on for centuries, even milleniums; I see no reason why it would change today
Tip of the day
In order to limit the hallucinations, it is absolutely necessary to limit the scope of the answers, so I put the following conditions before the actual questions:
āIn babylon.js, version 4.2, Javascript: Questionā
or
āIn babylon.js, version 5.5, Typescript: Questionā
The day when artificial intelligence will also have a conscience and when this artificial intelligence superior to man will realize how much man is a cancer for the planet by destroying it, well, we have collectively created artificial intelligence, but also push humans to extinction, unless global warming does it first.
I think we have to be careful how far we go with artificial intelligence. It can serve humans but also destroy them if this intelligence reaches higher levels of understanding of the world around it, of the human who destroys, pollutes, killsā¦ Will this artificial intelligence take measures to survive, or will he see himself as superior to humans.
Some people think itās great, but not so much. Why do you want to create intelligences that will ultimately be superior to humans. We put thorns in our feet.
Thatās just my opinion.
This is just āsci-fiā to me and VERY UNLIKELY to happen. The AI is just āour puppetā and can live only as long as it serves its āmasterā purpose and objectives. If it turns out to become a threat for this purpose and objective, it will be eradicated a lot faster than it took to create it. I have no fear about how the AI could act. My only fears are about the humans that (think) they control it, yet have no faen clue of the implications of what they are teaching it to do.
Anyone to tell how much of this answer has been āinjectedā in the core 
Of course not, because we have no visibility on this. 
And THIS is currently the one issue WE need to fix.
As always, my opinion only.
The same can be said of the human mind. This reminds me of a film where the human asks the machine if it can prove that it is sentient. The machine answers back with the same question: can you prove it of yourself?
I contend that by engaging with AI, one has never had a clearer mirror to recognize and understand oneself.
It is a gift to mankind!
1 Like
@mawa All the technology that humans create and inspired by science fiction at the base. Those who create science fiction films imagine this with their human minds based on what the future could be. Then others find it cool and say why not invent it. Some have a mind that is not necessarily well placed, some may want to use this AI as a weapon or add it to a weapon.
Watch humanoids create, some want to create robots that look like humans, think like humans. It exists and itās inspired by science fiction.
Do what I said is that if we add consciousness to AI, thatās where science fiction will become very real.
Look at cell phones for example, 30 years ago people also said it was science fiction, and yet it is created. So what is not probable, can become so.
Mobile phones, although useful, also create a lot of problems for future generations: less memory work, problem with fingers, people are a lot on their phones while driving, in restaurantsā¦ In short, it also does damage and we had not imagined all this.
AI will be the same, we think we control, we think we do things well, but we donāt think of everything. They say itās science fiction. But science fiction already exists in reality.
The human has created a lot of things that thought they could control.
1 Like