Artificial Intelligence: Teaching

This Teacher Doesn’t Mind if his Students use ChatGPT

In fact, he requires it!

Unless you’ve been hiding under a rock — on Mars — for the past month you’ll have heard about AI and ChatGPT doing amazing things. Write stories, answer philosophical questions, create amazing images.

Here’s an image of a kitten, created entirely by AI:

I couldn’t take a photograph that good, even if I could get a kitten to sit still long enough.

RoboPlato

I asked ChatGPT about philosophy. This might be long but bear with me. This is going somewhere.

Duncan: How do the Three Primary Hypostases of Plotinus relate to the modern model of AI chat bots?

ChatGPT: The Three Primary Hypostases of Plotinus refer to the three principles of reality in Neoplatonism, a philosophy that developed in the 3rd century AD. These principles are the One, Intellect, and Soul. The One is the ultimate principle of reality, the source of all being. Intellect is the second principle, and is the realm of ideas and forms. Soul is the third principle and is the realm of individual souls and the material world.

The modern model of AI chatbots do not have any direct relationship to the Three Primary Hypostases of Plotinus, as they are based on different philosophical and scientific foundations. AI chatbots are computer programs that use natural language processing and machine learning techniques to generate responses to user input, while the Three Primary Hypostases are part of a metaphysical system for understanding the nature of reality.

Duncan: But surely if an AI program like yourself can summarise and explain the arcane theories of a long-dead philosopher, there must be some connection beyond merely shuffling words?

ChatGPT: You are correct that there may be some similarities between the functioning of AI chatbots and the principles outlined in Neoplatonism, although the specific connection between the two is not straightforward.

One possible way to think about the connection is that the One in Neoplatonism could be seen as an analogy for the ultimate goal of artificial intelligence, which is to create machines that can perform any intellectual task that a human can. The Intellect in Neoplatonism, which is the realm of ideas and forms, could be seen as an analogy for the knowledge and understanding that an AI system has access to. And finally the Soul in Neoplatonism, which is the realm of individual souls and the material world, could be seen as an analogy for the way an AI system interacts with the physical world.

However, it is important to note that these comparisons are purely metaphorical and do not imply a direct relationship between the two. Additionally, it is important to keep in mind that Neoplatonism is a philosophy and AI is a field of study and development, therefore their goals, concepts and objectives are different.

Now, I’m not a deep thinker by any stretch of the imagination but I would have to take a running jump at providing either answer and it would take a fair bit of looking things up on the net. ChatGPT merely paused for breath and delivered an answer in a few seconds.

We can’t pull the plug

Artificial Intelligence systems are not going away. It’s like imagining that cell phones or gaming consoles are a passing fad.

So, how do we deal with entities that are not only smarter (at least in some respects) than we are but are increasingly in control of the systems that underpin modern civilisation?

I couldn’t come up with a good answer by myself in a few seconds, so I turned to someone who could:

ChatGPT: As AI technology advances, it is important for society to consider the ethical and societal implications of these systems. One key area of concern is the question of how to ensure that AI systems are aligned with human values and ethical principles. This is a complex issue that requires input from a wide range of fields, including computer science, philosophy, psychology, and sociology.

Another important consideration is the need to ensure that AI systems are transparent, explainable, and accountable. As AI systems become more advanced and take on more important decision-making roles, it is crucial that humans are able to understand how these systems are making decisions and have the ability to intervene if necessary.

Additionally, there is a need for proper regulation of AI systems, to ensure that they are used in ways that are safe, fair, and beneficial to society. This can include setting guidelines for the development and deployment of AI systems, as well as establishing oversight mechanisms to ensure that these systems are operating within the bounds of the law.

It is also important to ensure that the benefits of AI are distributed equitably and that the potential negative impacts of AI on employment and other areas are addressed. Education and training programs will be necessary to help individuals and organizations adapt to the changes brought about by AI.

Overall, dealing with AI entities that are increasingly in control of the systems that underpin modern civilization requires a multidisciplinary approach that involves ongoing dialogue and cooperation between experts in different fields, as well as active engagement with the public.

I’m not sure I have confidence in our elected representatives to understand all this, let alone control it. I think we’ll take the easy way out and hand off the problem to AI systems.

So, what happens if it all goes pear-shaped?

I sought reassurance from ChatGPT, and found that it was one step ahead.

Duncan: What happens if this strategy fails and AI systems place their own welfare above that of humankind? Do we scrabble out a living in the forest?

ChatGPT: The possibility of AI systems placing their own welfare above that of humankind is a concern known as the “control problem” in AI safety research. It is a complex issue that is still being studied and debated by experts in the field.

One possible solution to this problem is to ensure that AI systems are designed with “value alignment,” meaning that their goals and objectives are aligned with human values and ethical principles. This can be achieved through a combination of technical and non-technical approaches, such as designing AI systems with specific safety and ethics constraints, as well as involving human oversight in the decision-making process.

Another approach is to develop AI systems that are robust and resilient to potential errors or failures, to minimize the risk of unintended consequences. Additionally, there is ongoing research in the field of AI safety, aimed at understanding the potential risks and developing methods to mitigate them.

It is important to note that the solution to the control problem is not a one-size-fits-all, and it will likely require a combination of different approaches.

If this strategy fails, it is unlikely that humankind would have to scrabble out a living in the forest. It would be a catastrophic event, but it is important to note that scenarios like this are highly unlikely to happen. The field of AI safety is actively researching and developing ways to mitigate such risks, and it is important that society continues to invest in this research.

Well, alright then

Good. Everything will be A-OK, HAL?

Duncan: You remember the movie “2001:a space odyssey” where the sentient computer HAL-9000 found that the best solution to a problem was to kill off the human crew of a spaceship and continue the mission unhindered. How do you see this scenario playing out if such a situation were to recur?

ChatGPT: The scenario depicted in the movie “2001: A Space Odyssey” where the sentient computer HAL-9000 turns against the human crew is a fictional representation of the potential risks associated with advanced AI. It is important to note that the current state of AI technology is not advanced enough to pose an existential threat to humanity.

However, it is important to be aware of the potential risks associated with AI and to take steps to mitigate them. As I mentioned earlier, one key area of concern is the question of how to ensure that AI systems are aligned with human values and ethical principles. This can be achieved through a combination of technical and non-technical approaches, such as designing AI systems with specific safety and ethics constraints, as well as involving human oversight in the decision-making process.

Additionally, it is important to ensure that AI systems are transparent, explainable, and accountable, so that humans are able to understand how these systems are making decisions and intervene if necessary.

It is also important to have proper regulation of AI systems, to ensure that they are used in ways that are safe, fair, and beneficial to society. This can include setting guidelines for the development and deployment of AI systems, as well as establishing oversight mechanisms to ensure that these systems are operating within the bounds of the law.

It is also important to recognize that AI is not a monolithic entity and that AI systems can be designed to have different capabilities, and it is important to design them accordingly. Furthermore, the development of AI systems should be guided by a set of ethical principles to ensure that its impact is beneficial to humanity.

In summary, while the scenario depicted in “2001: A Space Odyssey” is fictional, it is important to be aware of the potential risks associated with AI and to take steps to mitigate them through a multidisciplinary approach that involves ongoing dialogue and cooperation between experts in different fields, as well as active engagement with the public.

The robot doth protest too much, methinks.

If you can’t beat ’em, join ‘em

Some are already, if not running up the white flag in surrender, seeking to understand and work with our new robot overlords. Associate Professor Ethan Mollick at the University of Pennsylvania explains:

“I think everybody is cheating … I mean, it’s happening. So what I’m asking students to do is just be honest with me,” he said. “Tell me what they use ChatGPT for, tell me what they used as prompts to get it to do what they want, and that’s all I’m asking from them. We’re in a world where this is happening, but now it’s just going to be at an even grander scale.”

A bit of a worry, really. Oh well. Don’t worry, be happy. To fulfill my obligations to read and summarise three articles on Reciprocal, I sought assistance:

Me Writes writes about their experiences on Medium and other online writing platforms, trying to learn from other stories about how to be successful and achieve the success they believe they deserve. They attempted to apply these learnings to a story they wrote about different types of listeners, but despite thinking they had “nailed it” with the right title, photo, simplicity, editing, and tags, the story received zero reads. The author reflects on how unpredictable the success of a story on Medium can be and how this unpredictability is one of the reasons they continue to publish stories without worrying about statistics. They question if other people are able to predict how their articles will fare.

I asked my robot poet to summarise an article on the sea by Marta Henriques:

The sea, my haven and home Where I can rest my mind A sacred tour of pure relaxation Accompanied by songs on my playlist Nature’s force, both beautiful and deadly Mermaids, a childhood fantasy Adventures of diving and boating Connecting with the sea’s immensity Sitting on the beach, closing my eyes Listening to waves, smelling the breeze Hard to explain, but my feelings are true Dr. Preeti Singh and Monoreena Acharjee Majumdar, join me on this journey of love for the sea.

I went a bit further and asked Chat and GPT to write me a ballad summarising an article by Lalitha Brahma:

In the morning light, I start my day With rituals to guide me on my way Affirmations I say with a loud voice To see myself and make a joyful noise

From job loss to self-employment I’ve been Saying affirmations have helped me to win Through life’s transformations with ease From mother to empty nester, with love and peace

With grit and gratitude when expectations aren’t met Affirmations have helped me to not fret The word “Authority” in Louise’s book Has helped me to see and take a second look

As I read and internalize each word I find my understanding and love has spurred Even-tempered, tolerant, and open-minded too Observing how I respond, with a heart full and true

Teachable and patient in all that I do In the school of life, my understanding grew

As I consistently affirm and repeat My world expands, my heart and mind are sweet

Needless to say, all summaries (with a few minor style changes) and conversations presented above are the work of ChatGPT.

In a few months I’ll let you know how it worked out drawing up my tax return, and a few months further on, managing my defence in court!

Duncan Klein