. 

“Artificial Intelligence”

by Micorsoft Image Creator from Designer

Powered By DALL-E 3

Happy New Year Everyone.

Looking back on the hot technologies of the world during 2023, it isn’t difficult to predict that 2024 will be a year in which the science, technology, and use of artificial intelligence (AI) is going to grow exponentially. Explodes sounds more like it. AI is expected to change our lives very soon, becoming more and more evident in this new year. Many of us still aren’t quite sure exactly what AI is yet, regardless of the level of education one has. As AI technology is beginning to assert itself as a major component of our personal and professional lives, along with it comes a lot of questions yet to be answered. Legal experts and regulators currently are finding themselves in new, untested territory.

AI isn’t new, of course, even though it kind of feels like it is. 2023 was the year that AI really blossomed. We have had AI at our fingertips for a few years now with our smartphones’ search capabilities, voice recognition and biometrics. But take a few steps back further, and we all have been using AI ever since we have been using calculators; that is, using a machine to do some of the thinking for us. An interesting article published by Tableau summarizes the history of AI. Click on the link to check it out.

The AI garnering the most attention these days is generative AI. Generative AI is a classification of what is known as machine learning, where a computer software program compiles all of the data entered into it, both textual and visual, assimilates relationships that reoccur in the data, and formulates a product, or answer, that is the most logical to it when queried. Just like a calculator will provide the answer of 4 when “asked” what is 2 + 2, generative AI will give you an answer to complex questions and requests, dependent upon the data it has stored.  One can ask generative AI to “create a poem about a dog and a chicken in Shakespearean prose” and, using all of the informational relationships programmed into the database it has access to, it will do just that. It can do the same with visuals – if you ask it to create a picture of a sunrise, it will come up with something photorealistic depicting a sunrise. Generative AI is what is known as a large language model (LLM), meaning it uses very large amounts of language-structured data to “learn” from during “training”.

As one can imagine, the quality and accuracy of the data the program was trained on is imperative to the usefulness of the program. The AI programs catching most of the attention now are the ones that are open source; in other words, accessing the data from the world wide web which anyone is capable of contributing to. This data set is humongous, as one could imagine. However, not all of it is automatically included in AI models, thank goodness. AI developers are capable of picking and choosing which sources are swept into its database. Great creations can be generated from all of this data, but there is a risk that it may come with inaccuracies and misinterpretations. Care must be taken when trusting the result. 

This area is also, rightfully, the focus of copyright law. Today, in fact, lawsuits are being filed by human content creators such as traditional media publishers, entertainers and other artists about “open source” being “fair use”. The courts are being asked to weigh in. This I am sure of, because I looked it up before starting the blog, and because it is important; anything created by artificial intelligence is not subject to copyright protection in and of itself. But asking artificial intelligence to produce something which mimics the work or likeness of another specific human being and claim it as your own is a no-no, as it should be. This still doesn’t get to the bottom of this. Is the above example of my artificially created Shakespearean poem about a dog and a chicken a no-no because I literally want to copy Shakespeare?

The complexities with AI can be seen when examining the course of IBM’s Watson. Watson was the first contemporary AI platform out of the gate, so to speak, when it came to offering language-based AI on a large, well-known scale. Watson was created to compete on the television quiz show Jeopardy! in 2011 and won! It was a natural language model, rather encyclopedic, prompted by keywords. Watson was not a resounding success in other uses, however. As is often the case, IBM was early with deploying newer technologies. Watsonx is their current version of LLM cloud-based AI and is very competitive with its peers in the industry. By the way, this fantastic American company offers educational training, not only in AI but in several technology specialties, and also offers plenty of job opportunities. 

Closed sources trained on proprietary input, in my opinion, is where the real horse race is going to be, and it is an open field right now. Trustworthy companies should be scrambling right now to assist business enterprises, education, the legal field, retail, the government, you name it, in developing secure in-house AI technologies. There is plenty of room for new businesses to competitively emerge in this arena as well. But this is an area is where it is crucial to get it right. This can’t be emphasized enough. It is crucial to get it right. Reputations are at stake, not to mention trusted productivity and outcomes. Many very intelligent people are warning there could be catastrophic consequences if it is not done right.

Generative AI is only one small subsector of artificial intelligence. Below is OpenAI’s response when asked to define artificial intelligence:

Artificial Intelligence (AI) is a branch of computer science that focuses on the development of systems and machines capable of performing tasks that typically require human intelligence. The goal of AI is to create machines that can simulate various aspects of human cognition, such as learning, problem-solving, perception, language understanding, and decision-making.

AI systems are designed to analyze and interpret data, adapt to changing circumstances, and improve their performance over time without explicit programming. These systems can operate in a wide range of domains, automating tasks, solving complex problems, and enhancing efficiency in various fields.

Key components and techniques within the field of AI include:

1. Machine Learning (ML): A subset of AI, machine learning involves the development of algorithms that enable machines to learn patterns and insights from data. Machine learning models can make predictions or decisions based on their training data and improve their accuracy as they encounter more information.
2. Natural Language Processing (NLP): NLP focuses on enabling machines to understand, interpret, and generate human language. This includes tasks such as language translation, sentiment analysis, and chatbot interactions.
3. Computer Vision: Computer vision allows machines to interpret and understand visual information from the world, including images and videos. Applications include facial recognition, object detection, and autonomous vehicles.
4. Expert Systems: These are AI systems designed to emulate the decision-making abilities of a human expert in a specific domain. They use knowledge bases and rule-based reasoning to solve problems.
5. Robotics: AI is often integrated into robotic systems to enable them to perceive their environment, make decisions, and perform tasks autonomously.
6. Speech Recognition: AI systems can be trained to recognize and understand human speech, enabling voice-activated interfaces and speech-to-text applications.

Artificial Intelligence is categorized into two main types:

1. Narrow AI (Weak AI): This type of AI is designed to perform a specific task or a narrow set of tasks. Examples include virtual personal assistants, recommendation systems, and image recognition.
2. General AI (Strong AI): General AI refers to machines with the ability to apply intelligence across a wide range of tasks, similar to human intelligence. This level of AI is theoretical and does not yet exist.

AI technologies are widely applied in various industries, including healthcare, finance, education, transportation, and entertainment. As AI continues to advance, it poses both opportunities and challenges, with ethical considerations regarding privacy, bias, transparency, and the societal impact of automation.

                   [OpenAI ChatGPT 3.5]

As you can see, the whole field of artificial intelligence is an open book, with no final chapter in sight and lots to talk about. We’ll be revisiting the subject occasionally here at Tomorrow Today, so please stay tuned.