Artificial Intelligence Primer

We are in the early stages of one the greatest technological revolutions of our time and I highly recommend taking time to understand the basics of it. Not many years from now, people will debate which had the greatest impact on mankind – the personal computer, the smart phone, the internet, or artificial intelligence (AI), more specifically artificial general intelligence, known as AGI or strong AI.

What is artificial intelligence? Basically, we have learned to teach computers to think, to learn, to plan, to converse, to create, and integrate these skills, the hallmarks of intelligence. The remaining skills are imagination (the ability to form novel mental images and concepts), and autonomy. With AI generating new content of all kinds, one wonders whether artificial intelligence has already achieved imagination or just new ways of combining the myriads of information with which it was programmed.

More human traits such as self-awareness, sentience (feelings) and consciousness seem to currently be beyond its grasp and may always be, though there is debate about that.

Weird as it sounds, the day when you will not be able to tell whether you are talking to a computer or a person is now here. AI-enabled apps can hold a conversation, solve problems, do research, and make recommendations tailored to you.

AI is also creating art, from photographs to paintings to sculptures. AI is not only writing term papers, but research papers, and even songs. I downloaded a free AI app called Splash Music and asked it to make up a song using the phrase, “Jesus, I am so thankful.” It made one up in 15 seconds with lyrics beyond what I requested and played it back as a 15-second rap. If I had signed up for a $15 subscription it would have sung it to me with its own created melody and words based on my request.

Click to play. You may need to tell your computer to open with Media Player.

I also downloaded a free app to my computer called ChatGPT and asked it to write a sermon based on 1 Cor. 13:5 – “love does not seek its own.” In less than 30 seconds it wrote an entire sermon.

I then asked it to write a romantic sonnet for my wife and again, I had one in a few seconds. Next, I asked it not to use passive sentences or flowery language and it wrote another version and will likely keep those stylistic instructions for whatever else I ask it to write, in other words, learning my preferences. Here’s what it did.

In the quiet moments, where words may falter,

Your presence, my anchor, my life’s sweet alter.

With steady grace, through joy and stormy weather,

We navigate this life, hand in hand, together.

In your laughter, a comfort, a soothing song,

In silence and chaos, where we both belong.

Through seasons passed and those yet to unfold,

Our shared journey, a story, tenderly told.

 With every shared glance, a world unfurls,

In the tapestry of us, life’s precious pearls.

No grandiose verses, just a truth profound,

In the simplicity of us, forever bound.

This attention to my instructions regarding style is a basic concept of AI – it learns and adapts. This is why people are excited about soon having a personal assistant, otherwise known as an “agent” who learns who they are, what they do, how they write, how they research, their habits, vocabulary, preferences, online experiences, shopping habits, and will become a not-quite-human, intelligent, thinking, learning, extremely capable personal assistant to do all sorts of things.

One thing to remember is that AI apps as programs are creations of a team. So, the viewpoints, beliefs and biases (true or false), and information (true or not) all get programmed in as well. Perhaps like a child this all gets examined by the app as it learns and receives further training and customization from the end user and gets modified, reinforced, or discarded down the line. Just know that this is an area of concern, but it seems like the widely broadcast wrong answers from sometimes poor inputs in these early days are just growing pains, though they may always be a concern at some level.

According to ChatGPT after I prompted it to answer this, the most common uses of ChatGPT are:

  1. Content Generation:
    • Creating blog posts, articles, and other written content.
    • Drafting emails or messages.
  2. Programming Assistance:
    • Generating code snippets.
    • Assisting with coding-related queries.
  3. Language Translation:
    • Assisting with translation between languages.
  4. Tutoring and Learning:
    • Providing explanations and answering questions on various topics.
  5. Customer Support:
    • Handling basic customer queries and support requests.
  6. Creative Writing:
    • Generating ideas for stories, poetry, or other creative works.
  7. Conversation and Entertainment:
    • Engaging in casual conversation for entertainment purposes.
  8. Information Retrieval:
    • Answering factual questions and providing information on a wide range of topics.
  9. Simulating Characters in Games:
    • Creating interactive and dynamic characters in video games.
  10. Prototyping and Design:
    • Assisting in the ideation phase of product design and development.
  11. Medical Assistance:
    • Answering medical queries and providing information (with limitations).
  12. Language Understanding Research:
    • Supporting research in natural language processing and understanding.
  13. Legal Assistance:
    • Assisting with legal research and providing information on legal topics.
  14. Mental Health Support:
    • Offering emotional support and basic mental health guidance (with limitations).
  15. Simulation and Training:
    • Simulating characters or scenarios for training purposes.

AI is being taught how to do mundane tasks and in the very, very near future will do the mundane part of just about anyone’s job. If the job is totally mundane, like taking fast food orders, filling out forms, answering common customer service questions, giving standardized training, writing computer code and such, it may totally replace the job. Some estimate that up to 50% of all jobs could be done by AI at some point. It goes beyond mundane tasks too – AI is demonstrably much better at detecting and blocking malware and hacking, i.e., cybersecurity than the best programmers and hackers.

Artificial intelligence is going to dramatically change our world with breathtaking speed, making companies more productive and profitable and replacing a lot of job functions and jobs. According to an article in Business Insider, the jobs thought to be most at risk are thought to be:

  • tech (coders, programmers, software engineers, data analysts)
  • media jobs (advertising, content creation, technical writing, journalism)
  • legal jobs (paralegals, legal assistants)
  • market research analysts
  • teachers (!)
  • finance (analysts, personal financial advisors)
  • traders
  • graphic designers
  • accountants
  • customer service agents

From ChatGPT May Be Coming for Our Jobs

AI-enabled computers are much better than humans at some tasks, learn new information much faster, and can retain far more information than any human, rapidly extending the AI advantage over humans in such areas as developing products and technologies, for example in pharmaceutical and other medical research. Very soon, AI will provide health care to remote locations and poor countries via Zoom-like teleconferences where AI an agent interviews the patient, coming up with its own questions and searching a massive database for a diagnosis, even ordering tests as would a doctor.

It is this access to far more information that a human could ever memorize, the input of many bits of instruction, topped off by the ability to reason and continually learn from experience and other new inputs that makes artificial intelligence so powerful.

It is also what has some people so alarmed. Many years ago, movies like 2001 – a Space Odyssey and The Matrix featured AGI computers taking over control and harming us. While some still see that frightening potential, others see harm in what people with malicious intent could use AI to do. For example, researchers at MIT wanted to know if AI could generate a formula for creating pathogens capable of causing widespread harm a la COVID or worse, biological warfare. In just one hour, AI delivered the formulas for creating in a lab four different pathogens capable of causing a pandemic.

Don’t think that North Korea, China, Iran, and others are not working on the same idea.

Just last week, Sam Altman, the CEO of OpenAI, provider of ChatGPT and probably the premier AI company in the world, was ousted from his job by board members concerned that he was pushing ahead too fast with developing AI capabilities (known as acceleration) and not paying enough attention to safeguards. However, a couple days later he was reinstated with the help of Microsoft, the company’s biggest investor, overruling the board and preventing the departure of 90% of OpenAI’s employees who said, if he goes, we go.

Artificial intelligence programs have been adopted at an unbelievably fast pace, far, far faster than any previous technological advance like the personal computer, smart phone or internet. ChatGPT, the free personal AI app, has been downloaded by 100 million people in its only first two months. Bing and Google are already using AI in internet searches.

The vast majority of people will before long have their own close to human artificial companion, only it won’t be physical, it will be on their phone and computer, a virtual assistant something like Alexa on steroids, becoming more and more human, only faster, smarter, more capable than humans in many ways, conversing, and creating, eventually even making decisions. There will even be conversations about whether AI agents should have human rights. In fact, that was one of a series of ethical questions the Wall St. Journal asked recently (see link below).

Do you remember the commercials years ago that asked, “Is it real or is it Memorex?” They were advertising high quality audio tape but very soon a similar question will be very common, asked about many things – “Is it real or is it AI?”

Change is underway, and it is on a much faster track with far more effects than many people know. There is a lot at stake, and lost jobs, greater efficiencies and profits, supersmart teachers and researchers, a medical revolution, a quantum leap in consumer convenience, and as yet unanswered ethical concerns are just the beginning. Just thought you should know.

I recommend reading the following news articles on AI.

Is AI a Risk to Humanity?

AI is Creating Ethical Dilemmas- How Would You Answer These?

The Contradictions of Sam Altman, AI Crusader