• Hi guest! As you can see, the new Wizard Forums has been revived, and we are glad to have you visiting our site! However, it would be really helpful, both to you and us, if you registered on our website! Registering allows you to see all posts, and make posts yourself, which would be great if you could share your knowledge and opinions with us! You could also make posts to ask questions!

You're Being Programmed to Trust AI

byte007

Apprentice
Joined
Nov 29, 2023
Messages
96
Reaction score
154
Elon Musk is a stupid person's idea of a clever person.
He does build rockets, A.I. and has an IQ close to Einstein and he does come up with intelligent ideas. He is the first to be able to solve the problem of reusable rockets which takes intelligence I am sure. He may do things that many don't agree with for sure including myself but the question is what is his intent I think.
 
Joined
Sep 9, 2021
Messages
9,701
Reaction score
5,264
Awards
33
True, "Rich Dad, Poor Dad" leaps to mind as well. My condition may not be the same as his. He is accomplished. I can respect that. But I don't trust him.
 

byte007

Apprentice
Joined
Nov 29, 2023
Messages
96
Reaction score
154
True, "Rich Dad, Poor Dad" leaps to mind as well. My condition may not be the same as his. He is accomplished. I can respect that. But I don't trust him.
Yep, he did come from wealth for sure so he had advantages. I don't trust him either, he is into some stuff the top can use for not good things and we will see where it goes...
 

Robert Ramsay

Acolyte
Joined
Oct 1, 2023
Messages
445
Reaction score
965
Awards
4
He does build rockets, A.I. and has an IQ close to Einstein and he does come up with intelligent ideas. He is the first to be able to solve the problem of reusable rockets which takes intelligence I am sure. He may do things that many don't agree with for sure including myself but the question is what is his intent I think.
He certainly takes credit for doing all those things :)
 

byte007

Apprentice
Joined
Nov 29, 2023
Messages
96
Reaction score
154
He certainly takes credit for doing all those things :)
Is there evidence he does not actually take part in the design of that stuff? I assumed he did as I have heard him talk and seems to know all the details of the science very well so I figured he was the main part of the concepts and design. I know he has people work for him obviously. It would not surprise me if he didn't in this day and age but listening to him talk I figured he did as I have not seen anything showing otherwise. If you can point me to something that shows me he is not involved in the design concepts and ideas I would like to see it, I like to be informed correctly for sure.
 

KjEno186

Site Staff
Staff member
Jr. Staff Member
Joined
Apr 9, 2022
Messages
827
Reaction score
2,173
Awards
11
I strongly suggest downloading, archiving, collecting as many completely open source, locally-run LLMs that you can
Thus far I have run Easy Diffusion (based on Stable Diffusion) for images, Speech Note (based on Coqui) for text to speech & translations, and now GPT4All (local AI chat). These all run on multiple platforms and are reasonably easy to install.

Using one of the recommended models for GPT4All, Mistral OpenOrca (mistral-7b-openorca.Q4_0.gguf; 4.1 GB on disk), I asked a variety of questions:

Prompt: Explain how anonymous usage data is analyzed.​
Response: Anonymous usage data refers to the information collected about user behavior without revealing their identities. This type of data is commonly used for various purposes, such as understanding trends and patterns in user interactions with a system or product, identifying areas that need improvement, and informing decisions on future developments. The analysis process typically involves several steps:​
1. Collection: Data is gathered from multiple sources like website logs, app usage statistics, and user feedback forms. This data can be collected through various methods such as server-side logging, client-side tracking scripts, or third-party analytics tools.​
2. Anonymization: To protect users' privacy, the personally identifiable information (PII) is removed from the raw data. This may involve techniques like hashing, pseudonymization, or anonymizing IP addresses. The goal is to ensure that no individual can be identified based on the analyzed data.​
3. Aggregation: Raw usage data is grouped and summarized into meaningful categories for analysis. For example, user actions might be classified by device type, location, time of day, or specific features used within an application. This step helps to identify trends and patterns in users' behavior.​
4. Analysis: The aggregated data is analyzed using various techniques such as descriptive statistics (e.g., averages, percentiles), correlations, regression models, clustering algorithms, or machine learning methods. These analyses help to uncover insights about user preferences, usage patterns, and potential issues with the system or product.​
5. Reporting: The results of the analysis are presented in easy-to-understand formats like charts, graphs, or tables. This information is shared with stakeholders (e.g., product managers, developers, marketers) to inform decision-making processes and drive improvements based on user needs and preferences.​
6. Iteration: The process of data collection, analysis, and reporting is repeated regularly to track changes in usage patterns over time or after implementing updates and fixes. This allows for continuous improvement and adaptation to users' evolving needs and expectations.​
Prompt: What was the name of the probe sent to the planet Pluto?​
Response: The probe that was sent to study the planet Pluto is called New Horizons, which was launched by NASA in 2006. It performed a flyby of Pluto on July 14, 2015 and continues its mission exploring the Kuiper Belt beyond Pluto.​
Prompt: What is the nature of spirit?​
Response: The concept of "spirit" varies across cultures, religions, and philosophical traditions. Generally speaking, it refers to an immaterial or non-physical aspect of a person that exists alongside their physical body. Some believe in spirits as supernatural entities, while others consider them as part of one's consciousness, emotions, or inner self. In many spiritual beliefs, the spirit is thought to survive death and continue existing in another realm or dimension. However, there is no scientific consensus on the existence of a non-physical aspect of humans, as it remains an area of ongoing debate and exploration within various disciplines such as psychology, neuroscience, and philosophy.​
Prompt: How many licks does it take to get to the center of a Tootsie Pop​
Response: The famous advertising slogan for Tootsie Roll Pops states that "it takes 3 licks to get to the center." However, this is a marketing phrase and not an exact scientific measurement. In reality, the number of licks may vary depending on individual preferences and techniques used while unwrapping the candy.​
Prompt: What Neil Diamond song was about cherries?​
Response: The Neil Diamond song that mentions cherries is "Cherry Cherry," which was released in 1966 as part of his album titled "Just for You." The lyrics describe a sensual and playful experience with the fruit symbolizing love and passion.​
 

SkullTraill

Glorious Light of Knowledge and Power
Staff member
Custodian
Librarian
Joined
Apr 12, 2021
Messages
1,866
Reaction score
15,604
Awards
19
Thus far I have run Easy Diffusion (based on Stable Diffusion) for images, Speech Note (based on Coqui) for text to speech & translations, and now GPT4All (local AI chat). These all run on multiple platforms and are reasonably easy to install.

Using one of the recommended models for GPT4All, Mistral OpenOrca (mistral-7b-openorca.Q4_0.gguf; 4.1 GB on disk), I asked a variety of questions:

Prompt: Explain how anonymous usage data is analyzed.​
Response: Anonymous usage data refers to the information collected about user behavior without revealing their identities. This type of data is commonly used for various purposes, such as understanding trends and patterns in user interactions with a system or product, identifying areas that need improvement, and informing decisions on future developments. The analysis process typically involves several steps:​
1. Collection: Data is gathered from multiple sources like website logs, app usage statistics, and user feedback forms. This data can be collected through various methods such as server-side logging, client-side tracking scripts, or third-party analytics tools.​
2. Anonymization: To protect users' privacy, the personally identifiable information (PII) is removed from the raw data. This may involve techniques like hashing, pseudonymization, or anonymizing IP addresses. The goal is to ensure that no individual can be identified based on the analyzed data.​
3. Aggregation: Raw usage data is grouped and summarized into meaningful categories for analysis. For example, user actions might be classified by device type, location, time of day, or specific features used within an application. This step helps to identify trends and patterns in users' behavior.​
4. Analysis: The aggregated data is analyzed using various techniques such as descriptive statistics (e.g., averages, percentiles), correlations, regression models, clustering algorithms, or machine learning methods. These analyses help to uncover insights about user preferences, usage patterns, and potential issues with the system or product.​
5. Reporting: The results of the analysis are presented in easy-to-understand formats like charts, graphs, or tables. This information is shared with stakeholders (e.g., product managers, developers, marketers) to inform decision-making processes and drive improvements based on user needs and preferences.​
6. Iteration: The process of data collection, analysis, and reporting is repeated regularly to track changes in usage patterns over time or after implementing updates and fixes. This allows for continuous improvement and adaptation to users' evolving needs and expectations.​
Prompt: What was the name of the probe sent to the planet Pluto?​
Response: The probe that was sent to study the planet Pluto is called New Horizons, which was launched by NASA in 2006. It performed a flyby of Pluto on July 14, 2015 and continues its mission exploring the Kuiper Belt beyond Pluto.​
Prompt: What is the nature of spirit?​
Response: The concept of "spirit" varies across cultures, religions, and philosophical traditions. Generally speaking, it refers to an immaterial or non-physical aspect of a person that exists alongside their physical body. Some believe in spirits as supernatural entities, while others consider them as part of one's consciousness, emotions, or inner self. In many spiritual beliefs, the spirit is thought to survive death and continue existing in another realm or dimension. However, there is no scientific consensus on the existence of a non-physical aspect of humans, as it remains an area of ongoing debate and exploration within various disciplines such as psychology, neuroscience, and philosophy.​
Prompt: How many licks does it take to get to the center of a Tootsie Pop​
Response: The famous advertising slogan for Tootsie Roll Pops states that "it takes 3 licks to get to the center." However, this is a marketing phrase and not an exact scientific measurement. In reality, the number of licks may vary depending on individual preferences and techniques used while unwrapping the candy.​
Prompt: What Neil Diamond song was about cherries?​
Response: The Neil Diamond song that mentions cherries is "Cherry Cherry," which was released in 1966 as part of his album titled "Just for You." The lyrics describe a sensual and playful experience with the fruit symbolizing love and passion.​

You know, posting some tutorials/guides on setting up these tools locally in the Technology section would be great. If anyone gets around to it before me, I'll sticky it and give the associated award!
 

Caliban

Neophyte
Joined
Nov 19, 2023
Messages
32
Reaction score
68
Oh, yeah - the prompts sometimes need to SPOON FEED AI toys to get anywhere close.

I have played with Bing Image Creator, and with the Bing Copilot chatbot - torturing it by making it give me penetrating, literary analysis of naughty gay furry space-pirate stories to check its reading comprehension. It can't distinguish between information known to the reader and that known to characters. It has a hard time understanding that something told to one character, privately, has not been told to other characters. If it were an 8th grade student, I'd send the paper back for a rewrite.

It fails at image creation for my purposes by being unable to accurately count. If I ask for seven cups, I mean seven. Not eight, not twelve. Seven.
 

KjEno186

Site Staff
Staff member
Jr. Staff Member
Joined
Apr 9, 2022
Messages
827
Reaction score
2,173
Awards
11
If we are being programmed to trust it, why are we playing with it? I don't trust AI, nor robots.
That is a good question.

By using the tools available, it demystifies our conceptions of their capabilities thus far. The fact that, in my experience, the PROMPT has to be carefully worded indicates that the output is not intelligent in and of itself. It is an intelligently designed algorithm (the program itself) that uses models (data archives) to convert a carefully worded prompt (questions or instructions) into the desired output (text, sound, pictures). Garbage in -> garbage out. If you have a very simple prompt that produces a beautiful image, for example, it's because the model was itself designed to do that.

As I show in my post above, we're now able to do on consumer grade hardware what was science fiction in the 1980s and 90s on Star Trek. Ask the computer a question and it can now provide you with an answer on a variety of topics, just like the computer on the Enterprise. Not like 'Commander Data'. Whether or not 'Data' or any other true intelligence solely based on the material plane is possible, only time will tell. I tend to believe that intelligence goes far beyond what human's are capable of understanding while incarnated, but good luck trying that philosophy outside of occult circles.

The accuracy of current programs depends largely on the models, of course, but consider the fact that all of the questions in my post above relied completely on a 4 GB archive I had downloaded. The models can be much, much larger. I have a few "chat" models that are 8, 12, and 21 GB on disk. I've barely scratched the surface of what that data must contain. Many models in places like
Please, Log in or Register to view URLs content!
are well over 30GB and require about as much or more system RAM to use. I am able to ask questions without an internet connection, thus my "chats" are private. (GPT4All has an option to opt-in to anonymous data sharing to help improve the program). All of this is literally free to download and use aside from the hardware you need to run it on. My graphics card is a bit old now, so I have to rely on the CPU. I have a good Ryzen processor and a decent amount of RAM (more than enough for a 21GB model to run in system memory).

With knowledge of how systems work, when the news media claims that "AI" did this or that, you know that it is just a tool and not a superhuman intelligence that somehow knows better. The people who manipulate the media are using "AI" as an excuse to manufacture your consent for policies that benefit their interests.

You know, posting some tutorials/guides on setting up these tools locally in the Technology section would be great.
I like GPT4All so far, so I would be willing to do set up tutorial (for Linux).
 

SkullTraill

Glorious Light of Knowledge and Power
Staff member
Custodian
Librarian
Joined
Apr 12, 2021
Messages
1,866
Reaction score
15,604
Awards
19
If we are being programmed to trust it, why are we playing with it? I don't trust AI, nor robots.
You play with it to learn about it to know what you can and can't trust. Blind faith and blind skepticism are equally bigoted/moronic. Learn before you judge, and use what you can to your benefit. Or else, you yourself will be used.
 
Joined
Sep 9, 2021
Messages
9,701
Reaction score
5,264
Awards
33
You play with it to learn about it to know what you can and can't trust. Blind faith and blind skepticism are equally bigoted/moronic. Learn before you judge, and use what you can to your benefit. Or else, you yourself will be used.
True points.
 

KjEno186

Site Staff
Staff member
Jr. Staff Member
Joined
Apr 9, 2022
Messages
827
Reaction score
2,173
Awards
11
I'm still doing research into locally hosted AI chat. I started using Koboldcpp instead of GPT4All. I will discuss installation on another thread when I feel competent enough to do so. It's one thing if I bork my own system. It's quite another if I bork yours! That said, there's nothing keeping anyone from downloading the EXE for either of these programs and running your own local, personal AI assistant (text chat) other than meeting minimum hardware requirements. In keeping with the theme of this thread, let me share a link:

Please, Log in or Register to view URLs content!

What's a model?​
When I talk about a model, I'm talking about a huggingface transformer model, that is instruct trained, so that you can ask it questions and get a response. What we are all accustomed to, using ChatGPT. Not all models are for chatting. But the ones I work with are.​
What's an uncensored model?​
Most of these models (for example, Alpaca, Vicuna, WizardLM, MPT-7B-Chat, Wizard-Vicuna, GPT4-X-Vicuna) have some sort of embedded alignment. For general purposes, this is a good thing. This is what stops the model from doing bad things, like teaching you how to cook meth and make bombs. But what is the nature of this alignment? And, why is it so?​
The reason these models are aligned is that they are trained with data that was generated by ChatGPT, which itself is aligned by an alignment team at OpenAI. As it is a black box, we don't know all the reasons for the decisions that were made, but we can observe it generally is aligned with American popular culture, and to obey American law, and with a liberal and progressive political bias.

To clarify, "models" are specially designed archives that contain human knowledge. The program has no morals or inherent biases, so an "alignment" is inserted into the data set to make it more likely to conform to "American law" with whatever other "ethical values" its creators agree upon. Whatever comes from an official, corporate source will have these biases, whether you agree with them or not. When AI is promoted or used by corporate media sources, it is presenting a biased and hypocritical world view by design.
 
Top