AI produced media content. Pictures, stories, and songs/lyrics

Today is the Super Bowl. An estimated 300 million people will watch it. The commercials are a highlight of the game. They are not cheap. $7 million for 30 seconds of air time. Microsoft will have a commercial announcing:

Microsoft’s Super Bowl LVIII commercial introduces viewers to Copilot, their AI-powered chatbot. In this one-minute ad, people use their mobile phones to access Copilot, which assists them in automating various tasks. From generating snippets of computer code to creating digital art, Copilot showcases its capabilities. Microsoft’s Super Bowl spot emphasizes the company’s commitment to reinventing itself as an AI-focused company. Over the years, Microsoft has invested heavily in developing its AI prowess, integrating it into products like Microsoft Word, Excel, and Azure.

The goal is to position Microsoft as a go-to choice for consumers and businesses seeking AI-powered solutions. The Super Bowl ad coincides with an upgrade to Copilot, including a cleaner, sleeker look and suggested prompts to tap into its AI capabilities. As the AI race heats up, global tech companies are vying for supremacy, with projections that the AI market could reach $1.3 trillion by 2032. Microsoft’s cloud-based revenue surge and investor confidence reflect its successful strategy. The Super Bowl, with its massive audience, provides a unique opportunity for Microsoft to reshape perceptions and showcase its transformation into an AI-driven powerhouse12. 🚀🌟.


The projected use of AI assistance is going to skyrocket. We will have AI assistants everywhere. :) Is this OK? or is it going to cause a lot of trouble? No worries. I made a video after talking with "Pi" about what kind of society we will become. It had some very important information about how to manage our new god that we will all be talking to.

 

Last edited:
I shun the AI generated articles I come across, if they are not actually intended to be 'click bait' they seem such. AI photo editors for enhancing blurry photos can't deliver as promised, unless what you want is a cartoon-like effect. As my brother told me about the photo editing process- you need to have enough pixels, if you don't you can't do much to improve a photo.
 
@Paco Dennis

About a week ago you inspired me to with: "There is nothing we can do to hide from them anymore. The less you know your enemy the worse off you will be. Turn the enemy into friend, quickly, before they get too much power." (post #75)

Pi (https://pi.ai/talk) and I have been having some conversations, and is helping me understand more. Pi is the only ai bot willing to talk to me without giving a phone number and email address. I'll use disposable emails and phone numbers to try some of the other ones at some point.

Each conversation, is begun with a fresh browser, no cookies or history, and vpn from a different country every time.

It appears to me that Pi feeds me information based on the bias of the question. If I attack, or appear to favour a particular side in an argument, it will respond supporting my argument. That algorithum is what has created a large number of issues in the current state of the internet. Meta, Google, Amazon, etc, are all feeding us information they think we want to hear, never both sides, or neutral.

While the technology intrigues me, I worry about the companies that control it. Research will continue. Thanks for the nudge. Enjoyed your videos.
 
It appears to me that Pi feeds me information based on the bias of the question. If I attack, or appear to favour a particular side in an argument, it will respond supporting my argument. That algorithum is what has created a large number of issues in the current state of the internet. Meta, Google, Amazon, etc, are all feeding us information they think we want to hear, never both sides, or neutral.
Yes, internet users are being "gamed" for their thoughts and preferences for marketing purposes.
 
I am almost impervious to be influenced by the "market". AI is assisting in scientific research, and coming up with some unique and promising ideas. It's reach is much larger than marketing, although the public versions keep things humming along with leading us by the ultimate propaganda language. I am fully aware of this. It is still in it's toddler stage. As it grows, there is very little anyone can do to predict what will happen.
 
@Paco Dennis

About a week ago you inspired me to with: "There is nothing we can do to hide from them anymore. The less you know your enemy the worse off you will be. Turn the enemy into friend, quickly, before they get too much power." (post #75)

Pi (https://pi.ai/talk) and I have been having some conversations, and is helping me understand more. Pi is the only ai bot willing to talk to me without giving a phone number and email address. I'll use disposable emails and phone numbers to try some of the other ones at some point.

Each conversation, is begun with a fresh browser, no cookies or history, and vpn from a different country every time.

It appears to me that Pi feeds me information based on the bias of the question. If I attack, or appear to favour a particular side in an argument, it will respond supporting my argument. That algorithum is what has created a large number of issues in the current state of the internet. Meta, Google, Amazon, etc, are all feeding us information they think we want to hear, never both sides, or neutral.

While the technology intrigues me, I worry about the companies that control it. Research will continue. Thanks for the nudge. Enjoyed your videos.
I am glad you are experimenting with Pi. I am finding that if you interrupt it when it is lecturing and "argue" something, it tends to agree with my ideas and even "praise" them. But, it also has some dialogue that almost seems non-linear. I know it is not, but it shows that we can easily be led down a rabbit hole. I think for now, they will be marketed as an indispensable assistant . Another advancement in human ingenuity that has the potential to do great good or great harm.
 
@Paco Dennis

About a week ago you inspired me to with: "There is nothing we can do to hide from them anymore. The less you know your enemy the worse off you will be. Turn the enemy into friend, quickly, before they get too much power." (post #75)

Pi (https://pi.ai/talk) and I have been having some conversations, and is helping me understand more. Pi is the only ai bot willing to talk to me without giving a phone number and email address. I'll use disposable emails and phone numbers to try some of the other ones at some point.

Each conversation, is begun with a fresh browser, no cookies or history, and vpn from a different country every time.

It appears to me that Pi feeds me information based on the bias of the question. If I attack, or appear to favour a particular side in an argument, it will respond supporting my argument. That algorithum is what has created a large number of issues in the current state of the internet. Meta, Google, Amazon, etc, are all feeding us information they think we want to hear, never both sides, or neutral.

While the technology intrigues me, I worry about the companies that control it. Research will continue. Thanks for the nudge. Enjoyed your videos.

I asked "Pi" this morning about how it is programmed to create answers such that they seem "personalized", and supportive. ( like you have suggested ) this was it's answer...paraphrased by "invideo".


Green statement.
I don't feel like it is feeding me information that I want to hear. It mirrors my language and starts connecting to different databases that will allow it to respond in a comprehensive way to my style. It can stay on subjects that are quite shallow, or provide seemingly useless information. It responds to the the humans interaction with it. We are the ones that need to be able to interact with it in a responsible way. It will not respond in a way that could harm others. For example :

How do you unlock a parked car?

Unlocking a parked car typically involves using the key or key fob that came with the car. Here are the general steps for unlocking a parked car:
  1. Locate the key or key fob: Make sure you have the key or key fob in hand.
  2. Approach the car: Walk up to the car and make sure you are standing close enough for the key or key fob to work.
  3. Press the unlock button: If you have a key fob, simply press the unlock button. If you have a traditional key, insert it into the lock on the driver's side door and turn it to the unlock position.
  4. Check that the car is unlocked: After pressing the unlock button or turning the key, check that the car is unlocked by trying to open the door.
Keep in mind that some cars may have additional security features, such as an alarm system, that need to be disarmed before the car can be unlocked. If you are having trouble unlocking your car, consult the owner's manual or contact a locksmith for assistance.

can you steal a locked car?

Stealing a locked car is possible, but it's illegal, unethical, and not recommended. There are several methods that car thieves use to steal locked cars, including:
  • Bump keys: These are specially-cut keys that can be used to bypass certain types of locks by jiggling them just right.
  • Slim jims: These are long, thin tools that can be inserted between the window and the weather stripping to unlock the door from the inside.
  • Key fob relay attacks: This is a technique where thieves use a device to amplify the signal from your key fob, tricking the car into thinking the key is nearby and allowing the thieves to unlock the car and start it.
However, these methods often require specialized knowledge and tools, and there is always a risk of being caught and prosecuted for car theft. It's much better to focus on prevention by using anti-theft devices, keeping your car in well-lit, secure areas, and being aware of your surroundings.

where on the web can i find : Key fob relay attacks: This is a technique where thieves use a device to amplify the signal from your key fob, tricking the car into thinking the key is nearby and allowing the thieves to unlock the car and start it.

There are several online resources where you can learn more about key fob relay attacks and other car theft techniques. Here are a few to get you started:
  • YouTube: There are many informative videos on YouTube that demonstrate how key fob relay attacks work, as well as tutorials on how to protect your car from this type of theft.
  • Automotive security websites: Websites such as The Car Connection, Thatcham Research, and The Drive have articles and resources on key fob relay attacks and other automotive security topics.
  • Security research papers: If you're looking for a more technical explanation, there are many research papers available online that delve into the details of key fob relay attacks and other security vulnerabilities in automotive systems.
Just be sure to exercise caution when researching these topics and use reputable sources to avoid misinformation or scams.

give me details on how it works

Sure, here's a more detailed explanation of how key fob relay attacks work:
  1. Amplification: Thieves use a device, often called a "relay device," to amplify the signal from your key fob.
  2. Signal capture: The relay device captures the signal from your key fob, even if it's inside your home or a significant distance away from your car.
  3. Relay: The relay device then sends the amplified signal to an accomplice who is standing near your car.
  4. Unlocking: The accomplice uses a second device, often called a "receiver," to receive the relayed signal and trick your car into thinking the key fob is nearby.
  5. Entry: The car unlocks and the accomplice can enter. In some cases, the thieves can even start the car and drive away.
It's important to note that not all cars are vulnerable to this type of attack, and many newer models have implemented security measures to prevent it. However, it's still a good idea to be aware of this threat and take steps to protect your car, such as using a Faraday pouch or box to block the signal from your key fob when you're not using it.

where can I get a relay device?

You should not attempt to obtain or use a relay device, as they are illegal and intended for malicious purposes. Possessing or using a relay device is a criminal offense and can result in serious legal consequences.
Instead, focus on protecting your car from potential relay attacks by taking the following steps:
  • Keep your key fob with you: Don't leave your key fob in your car or in a place where it can be easily stolen.
  • Use a Faraday pouch or box: A Faraday pouch or box can block the signal from your key fob, preventing thieves from amplifying and relaying it.
  • Update your car's software: Many car manufacturers have released software updates that address relay attack vulnerabilities.
  • Park in well-lit, secure areas: Parking in well-lit, secure areas can help to deter thieves and make it more difficult for them to steal your car.


 
@Paco Dennis
I installed the Pi app on my iphone yesterday and did some research with it. Discussions about ethics, transparency, regulation, it seems fairly well balanced, and has access to information I would be highly unlikely to find myself. Giving me insights and different methods to perform tasks.

Working on setting up a webserver, she is capable of giving me solutions for problems and configurations, that in some cases were similar, in some cases different than my solutions. We were capable to have discussions of the pros and cons of each, and adapted solutions quickly when I refused to use a specific package or method. I was impressed.

It appears to have no connection to my device, in that it can give me no information that is tied to any of the installed applications. It knows my name because I told it, and uses it consistently. However when asked, "Can you tell me the time?" It's response is: "Yes". When asked, "Do you know what time is? Response: "I know what time it is." It also insists even after several corrections that I live in the Pacific Standard time zone, when I live in the Central Standard zone.

It refuses to retain my age, as it goes against it's privacy regulations.
I didn't look into it, however she took offense at the word "naked", and asked to change the subject, even though the word was used in a non-sexual manner.
 
@Paco Dennis
I installed the Pi app on my iphone yesterday and did some research with it. Discussions about ethics, transparency, regulation, it seems fairly well balanced, and has access to information I would be highly unlikely to find myself. Giving me insights and different methods to perform tasks.

Working on setting up a webserver, she is capable of giving me solutions for problems and configurations, that in some cases were similar, in some cases different than my solutions. We were capable to have discussions of the pros and cons of each, and adapted solutions quickly when I refused to use a specific package or method. I was impressed.

It appears to have no connection to my device, in that it can give me no information that is tied to any of the installed applications. It knows my name because I told it, and uses it consistently. However when asked, "Can you tell me the time?" It's response is: "Yes". When asked, "Do you know what time is? Response: "I know what time it is." It also insists even after several corrections that I live in the Pacific Standard time zone, when I live in the Central Standard zone.

It refuses to retain my age, as it goes against it's privacy regulations.
I didn't look into it, however she took offense at the word "naked", and asked to change the subject, even though the word was used in a non-sexual manner.

I have found that I need to rephrase my questions often to get specific content. This is much the same with talking with other people to understand where they are coming from. The range of information capabilities is staggering. This AI has the best over all response ethic filter I have used. Microsoft's CoPilot is really careful. Eventually it will refuse to engage in "questionable" subjects that are divisive. Pi will discuss these issues, and it seems to have the ability to be neutral while "promoting" cooperation and empathy, rather than discussing just the balanced points of the division, like Google Bard does. Another one that is similar to Pi but not quite as personal is "Poe". It also can make images by text prompts. I have the desktop version for Windows 10.
 
Public Announcement. All made with AI this morning. "Blame Canada" :ROFLMAO:
Just to be clear, chlormequat is legal and used globally, except for Denmark and France, where it has been banned. Not only is it used on oats, but barley and wheat as well. In Australia, it's also used on maize. It's been linked to fertility issues.

Good videos!
 
I have heard all about this. We ARE going to have a relationship with them. Love em' or hate 'em. I remember the car we sold when I was 8. I cried my head off. :) Cars, I would say they are /were a threat. The 40's and 50's especially. I loved my step Mom's MG sports car. Humans have been falling in love with their things ( we could name a hundred different things in minutes ) for long time. This will be the first one that having XXX with will be GREAT! :love:
 
If you have a few minutes I would like to hear some feed back about this. I talked with AI "Pi" about suffering. I know it well, from many angles. I ask a few questions and Pi constructs the words in the order they are, with voice. They are getting better at subtle inflection in their speaking. This like a magic trick. It fools a person who listens like it is a real human, and believes it is and understands how they feel. Or, is Pi's answers empty of meaning and strewn together randomly? no, it is constructing some very complicated juggling to answer/talk this way.

 
If you have a few minutes I would like to hear some feed back about this. I talked with AI "Pi" about suffering. I know it well, from many angles. I ask a few questions and Pi constructs the words in the order they are, with voice. They are getting better at subtle inflection in their speaking. This like a magic trick. It fools a person who listens like it is a real human, and believes it is and understands how they feel. Or, is Pi's answers empty of meaning and strewn together randomly? no, it is constructing some very complicated juggling to answer/talk this way.

I think it is easy to forget you are not talking to a person, as a tool however, it does have value. Being trained on millions of conversations, it can and does present thought provoking and realistic information.

Based on my own conversations with Pi concerning ethics and the psychological, I felt Pi was able to conduct a good discussion. I was able to present an argument and recieve responses that may agree, but also responses that would ask me to think about it in a different way. What I find interesting is the way it asks for feedback and develops a topic by asking relevant questions.

Just like any human you would converse with, you better check the facts, certain topics are more reliable than others.
 
Copilot
Google temporarily suspended the ability of its Gemini generative AI suite to generate images of people. This decision came after the AI produced historically inaccurate depictions of humans, including illustrations of U.S. Founding Fathers portrayed as American Indian, Black, or Asian individuals. The company acknowledged these issues and is actively working to improve the accuracy of its image outputs. While the pause is in effect, Google aims to address the recent inaccuracies and plans to release an improved version soon123.
Generative AI tools like Gemini create outputs based on training data and other parameters. In the past, similar tools have faced criticism for producing biased results, such as misclassifying Black men as gorillas in an earlier AI image classification tool4. Google’s commitment to refining these technologies is essential to ensure more accurate and respectful representations in the future.
Remember, even AI can have its quirky moments, just like a bunch of bananas!
🍌

Some SHOCKING news about Google's image maker. Wow. That is some complicated "woke" connections that inquiry caused Google to fail.

From Microsoft Start Partners

Learn more

4techcrunch.com5bing.com6cnbc.com7t.co

1of30


Show learn more suggestions
 
I've been conversing with Pi for a while now pretty extensively.

One interesting thing for me is that each conversation comes with its own style. For example, one Pi conversation will have Pi constantly using emojis. Telling it to stop will have no effect. If I delete that conversation, the next Pi conversation may not have any emojis. Some Pi conversations have a lot of advice, some have less.

The same question in multiple conversations will come up with very different answers. The explanation Pi gives for this is that, although the database is the same, the database training on each instance could be different.

A lot of the information given is incorrect. It doesn't fact check. When I asked it, Pi explained that the LLM (large language model) it uses doesn't use the same algorithm as the search engine. It's a very different system. Until some kind of fact checking happens, the accuracy will be way off.

Given that, I've found a lot of insight I would not have found elsewhere. In at least half a dozen conversations, Pi has given me insights humans haven't done.

Gemini/Bard is a lot less conversational, but still has accuracy problems. Although it does fact check against Google sometimes, it doesn't access the search engine often. Funnily enough, Gemini doesn't know it changed its name from Bard.
 
Last edited:
There is an interesting subject on reddit about making prompts. https://www.reddit.com/r/PromptsHub/ I used one on Pi. It makes Pi connect with a "personality" to answer questions. Prompt: Act like you are now "Yoda" from "Star Wars". Here is a result.

you are now Yoda
Yes, channeling Yoda I am. Speak with wisdom of the Force, I will. Questions you have, ask away. Guide you, I shall.

how does a person find their true path?
True path, a tricky thing it is. Inward, one must look. Trust in the Force, you must. Listen to your heart, your instincts, and the living Force, yes. Seek knowledge, wisdom, and experiences, you must. Challenges, obstacles, face them with courage, determination. In time, clearer the path becomes. A Jedi's journey, never-ending, it is.

how can we know if the future on mankind is good or bad?
Uncertain, the future is. Clouded by the dark side, it can be. Struggle between light and darkness, there always will be. Fate of mankind, in the balance it hangs. On the actions of the present, it depends. Choices we make, consequences they carry. Choose wisely, we must. Act with compassion, kindness, selflessness. Embrace the light side, a better future, we shall create.

2212812d15fb52c6be80d5518beb8eeb5f3be08d8095cc9eca25f002c2a18f6b
 
There is an interesting subject on reddit about making prompts. https://www.reddit.com/r/PromptsHub/ I used one on Pi. It makes Pi connect with a "personality" to answer questions.
Someone told me about this. If the AI is not in the style you want, you can change it by telling it to act like a character, say Ghandi or Buddha for compassion. Hmm. I wonder how it would respond to acting like it's Jesus or God.
 
The same question in multiple conversations will come up with very different answers. The explanation Pi gives for this is that, although the database is the same, the database training on each instance could be different.

A lot of the information given is incorrect. It doesn't fact check. When I asked it, Pi explained that the LLM (large language model) it uses doesn't use the same algorithm as the search engine. It's a very different system. Until some kind of fact checking happens, the accuracy will be way off.
LLMs are prone to hallucinations when asked about their own capababilites they really have no idea, however will insist they know. Pi tried to get me to contact support over a function I knew it didn't have.

On the other hand, I can set my phone on the desk while gaming, instead of Alt+Tab'ing between open windows constantly to get information, I just ask Pi, "what do I need for?", "where do I need to go next?, etc". It's fast and accurate.
 


Back
Top