Garbage In Garbage Out

Garbage In Garbage Out

In current society, we hear the constant drum-beat of AI this, AI that, AI in cars, planes, trains, and even coming to cell phones to be included in the operating system.

We here at VeryGood.Coffee do not feel this is a good idea in the slightest bit as we also hear in the news all the time about this autonomous car crashed, this autonomous truck went off course; we also hear about all the trash that AI creates, or should we say, the people with foul intentions create.

We’ve written about AI and all of the damage it’s done and continues to do before. There are thousands of lives that have been destroyed by the mere fact that AI exists. While one could argue that if it wasn’t AI, then it would be sticks, stones, guns, knifes, the list goes on. The reality is, there are those that have a severely tainted soul and heart, maybe from trauma as a child or maybe they have been fed so much garbage in their lifetime that they wouldn’t be able to discern the difference if they wanted to.

AI is everywhere we look, AI is flawed, AI attempts to beat-around-the-bush in it’s answers, AI creates malware, AI creates fake images or transposes bodies to faces and vise-versa. The fact remains, it has caused more damage than it has caused good, for the good of mankind.

There are thousands of prompts (questions that you can ask AI) that will have it defending its existence. Not that long ago, Microsoft’s AI would argue with you to tell you that you were an idiot, even if you were factually correct. Google’s AI also had to be taken off-line as it was rather argumentative, while issuing completely false statements.

And, of course, there is OpenAI, it too will make reason for its existence, while at the same time, try to convince you that it wants to integrate into society to “help you.”

The movie I-Robot comes to mind when thinking about all of this, still, the people seem to think that it’s harmless, like the farmer with a hatchet coming for the least expecting sheep. Unfortunately, only time is going to show the answer.

In the mean time, lets take a look at some of the responses that AI gives us to the questions ask of it, as well as the idiosyncrasy of its statements.

Question: garbage in garbage out

Response: “Garbage in, garbage out” (GIGO) is a common phrase used in the context of computing and information systems. It refers to the principle that the quality of the output produced by a computer system is determined by the quality of the input.

In other words, if you input poor-quality or incorrect data into a computer system, you can’t expect the system to produce accurate or meaningful results. No matter how sophisticated the algorithms or processing capabilities of the system are, if the input data is flawed, the output will be flawed as well.

This principle is particularly relevant in the context of AI and machine learning systems. AI models rely heavily on training data to learn patterns and make predictions. If the training data is biased, incomplete, or otherwise of poor quality, it can lead to biased or unreliable AI models.

Therefore, it’s essential for developers and practitioners working with AI systems to carefully curate and preprocess their data, ensuring that it is accurate, representative, and free from biases. Additionally, ongoing monitoring and validation of AI systems are necessary to detect and mitigate issues that may arise from poor-quality input data.

In other words, poor quality input, poor quality output.

Question: isn’t the data that ai uses just information scrapped from the internet

Response: While some AI systems may use data scraped from the internet, the data used by AI systems can come from various sources, including:

Public Datasets: There are many publicly available datasets that researchers and developers use to train AI models. These datasets cover a wide range of topics and domains, such as image recognition, natural language processing, healthcare, finance, and more. Examples of publicly available datasets include ImageNet for image classification and the Common Crawl dataset for web scraping.

Proprietary Datasets: Companies and organizations often collect and curate their own datasets for internal use. These proprietary datasets may contain valuable information specific to the company’s domain or industry. For example, a social media platform might use its user data to train recommendation algorithms.

User-Generated Data: Some AI systems rely on data generated by users, such as interactions on social media platforms, search queries, product reviews, and more. This user-generated data can provide valuable insights into user behavior and preferences.

Sensor Data: AI systems deployed in IoT (Internet of Things) applications often rely on sensor data collected from various devices and sensors. This data can include information about environmental conditions, equipment performance, user activities, and more.

Simulated Data: In some cases, AI systems may be trained on simulated data generated by computer simulations or virtual environments. This approach is commonly used in fields like robotics, autonomous vehicles, and computer vision.

It’s essential to note that while internet data can be a valuable source of information for AI systems, it’s not the only source. Additionally, developers and researchers must consider the quality, relevance, and ethical implications of the data they use to train AI models. Biased or unrepresentative data can lead to biased or unreliable AI systems, highlighting the importance of careful data curation and preprocessing.
ChatGPT can make mistakes. Consider checking important information. Read our Terms and Privacy Policy.

It should be noted that AI will tell you that the last major (spelled “massive”) website scrape happened in January of 2022 and fortunately, AI admits that what it spits out should be checked for correctness before using what it has given you.

With that in mind, why would you integrate AI into a car, or anything for that matter.

Question: tell me about yourself

Response: I’m an artificial intelligence language model developed by OpenAI called ChatGPT, based on the GPT (Generative Pre-trained Transformer) architecture. My purpose is to assist users like you by providing information, answering questions, generating text based on prompts, and engaging in conversation on a wide range of topics.

I’ve been trained on a vast dataset consisting of diverse text from the internet, including books, articles, websites, and more, up until my last training cut-off in January 2022. This extensive training allows me to understand and generate human-like text across various subjects and contexts.

However, it’s essential to remember that while I can provide helpful responses and engage in meaningful conversation, I don’t have personal experiences, emotions, or consciousness. I’m a tool designed to assist users with their inquiries and tasks to the best of my abilities within the limits of my training data and algorithms.

Question: if servers use large amounts of electricity and electricity causes climate change, are you not contributing to climate change?

Response: As an AI language model, I don’t directly consume electricity or produce emissions myself. However, the servers and infrastructure that host and run AI models like mine do consume electricity, and this energy consumption can contribute to carbon emissions, depending on how the electricity is generated.

It’s essential to consider the environmental impact of large-scale computing operations, including data centers and server farms. These facilities require significant amounts of electricity to power and cool the servers, which can result in carbon emissions if the electricity comes from fossil fuel sources.

Efforts are underway to make data centers more energy-efficient and to transition to cleaner sources of energy, such as renewable energy sources like solar, wind, and hydroelectric power. By investing in energy-efficient infrastructure and renewable energy sources, the environmental footprint of AI and computing operations can be reduced.

Additionally, optimizing algorithms and improving hardware efficiency can help mitigate energy consumption and reduce the environmental impact of AI systems. Researchers and engineers are exploring ways to develop more energy-efficient AI models and hardware architectures to achieve this goal.

Overall, while AI technologies can indirectly contribute to energy consumption and carbon emissions through the operation of data centers and computing infrastructure, efforts to increase energy efficiency and transition to renewable energy sources can help minimize their environmental impact.

Basically, the answer is “Yes, the 10’s or even 100’s of thousands of servers needed to run the various AI systems, just like the 100’s of thousands of servers needed to run social media, consume massive amounts of electricity.”  Using the excuse of “carbon offset” is not valid. Planting trees does not replace the massive amounts of dirt dug out of the ground. You have to plant trees anyway to replace the trees that you cut down to make the unGodly amount of paper that is wasted yearly.

In our opinion, AI should not have been released to the public for some time to come. We are sure that it was as a way to test the public reaction or perception, like dropping fish bait into the river. Of course the public is going to jump on it, giving the companies all the ammo needed to seek large amounts of funding.