ChatGPT And A Dream

Two Pills

We’re at a moment that is oddly reminiscent of a scene in the 1999 classic The Matrix where Neo sits down in a high back leather chair across from a man he’s never met in a long black leather jacket and sunglasses. That man, Morpheus, says to Neo, “I imagine right now you feel like Alice, tumbling down the rabbit hole.” As the scene progresses, Morpheus sits down in the other chair he pulls something out of a box he had in his pocket, then opens both hands to Neo presenting a blue and a red pill. Morpheus presents Neo with a choice, “This is your last chance,” he says, “after this, there is no turning back. You take the blue pill and the story ends. You wake up in your bed and believe whatever you want to believe. You take the red pill, you stay in Wonderland, and I show you just how deep the rabbit hole goes.”

If you’re a blue pill person, this may be as far as you want to read. If you’re a red pill person, read on. 

GPT4

On March 14, 2023, OpenAI released a product called ChatGPT4. GPT4 is what’s called a Large Language Model (LLM). What is a LLM? It is a recent development in AI research. I could tell you about what it is, but these guys do it better:

I’m not going to sit here and pretend that I understand how this stuff works though, so I want to focus on how I have been using this technology so far. I’ll share some of my thoughts at the end, but what I really want is for you to do your own investigation and draw your own conclusions.


Most people that I talk to seem to have a very limited understanding of what modern AI is. I don’t blame them. Some of the most significant breakthroughs have only been happening in the last couple months and now they are coming so fast that you have to spend hours every week just to stay on top of what’s new.

The people I talk to about this stuff generally seem to fall into two camps.

The first camp is people who have heard of technologies like generative AI such ChatGPT or Midjourney and have this perception about it like its kind of cute, maybe gimmicky, and nothing really to worry about. For whatever reason, it’s pretty difficult convincing those people otherwise.

The second camp is people who have not heard of these technologies, and its a little bit easier for them to get it. 

It almost feels like there is a massive elephant in the room, but most of us are treating it like a dog. We play with it and teach it tricks not realizing how its about to trample our way of life.

My Experimentation

It can be difficult to know where to begin exploring with a new technology such as this. I want to tell you some of the ways I have used it so far. 


As I am sure, it is abundantly clear to most of the people who know me, I would like to move my life more in the direction of professional photography. One of the things I find overwhelming, is the sort of social media trap we’re stuck in where there is this culture of content creation. I am sure there are other ways to advertise yourself, but it also makes sense to try to reach people where they are, and the format for that in today’s world is short form video content. It can be a very daunting task to come up with ideas for content on the kind of schedule you need to post on, so I decided to ask GPT for help. 

It provided me with a list of ideas, and while some of them I didn’t care for, some did catch my attention.

Okay, if you know me, you would probably think that this is not really my cup of tea, however one of the suggestions did interest me. Point 3 on using photography to connect with nature. After having a bit of back and forth discussing some of the ideas it came up with (yes, I mean discussing), I came back to that topic.

Is this cheating? I don’t really know. People who have been at this for a while and can afford a team probably just hire somebody to do this for them. The point is that after playing around with GPT for less than an hour, I can come up with a year’s worth of content creation ideas, complete with scripts and all I have to do is put them together. If I am interested in ideas relating to a specific topic, I can just ask “how about some ideas relating to dealing with the unexpected?”  or “how about some ideas that will encourage interaction with my audience?” I have to tell you, many of them are pretty good.


Does this kill creativity? Not really. I suppose it comes down to the extent that you want to use it and how involved you want to be in the creative process. In the example above, I asked it to write me a script so I can fulfill a marketing task of churning out content so that social media algorithms will push my work to a wider audience. In this case, I would rather spend my time and creative energy on making art. 

In another example, I asked GPT to help me put together a business plan. Following up, I asked it to help me figure out my customer avatar. It asked me a bunch of questions about the demographics, values, interests, and other such things of my ideal customer. I had to provide that information myself, but GPT was definitely a useful guide through the process. 

Later on once we had determined who my ideal customer was, I asked GPT to provide a list of other people offering the same service to the same customer. At first it provided me with a list of YouTubers, so I asked GPT to redo the list, with more of a focus on photographers. It provided a list of several people. Some of them I’d heard of, some of them I hadn’t. I asked it to add a few more to which it happily obliged, and then I decided to take things a level deeper.

I asked it to tell me in general terms what those people did well and what they could do better. Then I asked if I were to offer that service to that audience, how might I improve the offer based on the things that those people don’t do well. 

Try It Yourself

I hope by now you’re starting to get a sense of what GPT is capable of. I have found that a fun exercise is to ask GPT to come up with a list of questions you can ask it to demonstrate its critical reasoning capabilities. Some of the questions are pretty insightful. The answers moreso. Don’t take my word for it, try this exercise for yourself. It is important that you gain first hand experience dealing with this technology. Don’t stop there either. Ask GPT follow up questions based on the answers it provided you. 

Here are some questions to get your creative juices flowing on ways to use this tool:

  • Help me come up with 10 project ideas for X

  • I have taken a photo of <describe your photo>, help me come up with an Instagram caption

  • I am considering <describe your idea>, please play devil’s advocate for me

  • I have options X and Y, what are some pros and cons of each?

  • I have these food items to use up, help me come up with a recipe using these ingredients, and other common kitchen ingredients

    • Provide the nutrional information for this meal

  • I am preparing for an interview for a company that does <what the company does>, what are some questions they are likely to ask me

    • Evaluate my answers

  • I am presenting my idea _____ to ______ who I believe has a different perspective than me because ______. What are some ideas for presenting my idea in a way that will resonate better with them?

  • Help me plan an event for ________.

  • This plan for _______. Am I missing anything?

  • What are some cool date ideas in <city>?

  • I want to learn <subject>. Please act as an expert tutor. I will provide you with the subject, material preference, and time I have available to study. You will then write me a study plan.

  • How will AI impact the business of photography in the future?

    • How can take advantage of being an early adopter of the technology?

    • What are some business ideas using AI for photography?

    • What are some criteria for evaluating those ideas as a business?

    • If there was a total score of 100, provide a weight to those criteria.

    • Evaluate this idea.

    • Provide the reasons for that evaluation

  • Brainstorm some ideas that I can use to connect socially with more humans

These are only ideas to get you started. If you work through some of them yourself, you will quickly come to the realization how powerful this tool is. Keep in mind, that you can and should ask it follow up questions. Sometimes, you will need to use several prompts to set up the context for the questions you want to ask it. You may need to even prime it with examples. In one case, I asked it to write a blog post for me (not one I shared anywhere) in my own style. I provided it with the text from two posts I’d written in the past so it could analyze my writing and write something new as if I did it. 

Taking It Further

If you’re still with me, I hope you’re starting to become pretty impressed by this tool. I also hope that you’re inspired to try it for yourself. 

Before I move on, I want to give one more example of how I have used this tool. I watched a YouTube video a couple weeks ago about how some guy used GPT to write the code for a simple side scroller game. I was determined to give it a shot. I told GPT that Iwanted to create an app, but had no experience with writing code. We had a bit of a conversation about what the app would be for, and what platforms it would be on. GPT recommended two coding languages, but helped me narrow it down to one, based on my skill level and app requirements. It then walked me through installing the software I needed, including troubleshooting that Google and YouTube were seemingly unable to help me with. 

I told it about the app I wanted to make. It provided me with suggestions for the user interface. It helped me solve problems along the way. It wrote all of the code. It was even able to walk me through the user experience of features we hadn’t developed yet (though we talked about them) and understand the intention of the app from a technical description. Yes, it comprehends my intentions. It interpreted the data I provided and drew a correct conclusion about how the app was meant to be used. 

While this particular app won’t work out, GPT did everything I asked it to. The problem is with my design. I am also confident that between GPT and I, we can eventually find a solution.

The Bigger Conversation

What I hope you understand so far, is that we all have access to this super powerful tool. You’re probably getting the sense that this thing can change the world. I started with what I have used GPT for so far because I want you to believe me when I tell you how amazing this is. Like I mentioned previously, don’t just take my word for it. Experience this for yourself. 


It is important that you believe me, because what comes next can be very disturbing. We are all in for a tremendous amount of shock. The world as we know will never be the same. This will happen much sooner than we think. This is not a conspiracy theory. This is not a gimmick. This is the real deal. 


From what I am able to tell, there needs to be a multi-dimensional conversation about this technology and we ALL need to be talking about it. 

  1. Small scale good stuff (We just talked about it)

  2. Big scale good stuff

  3. Small scale bad stuff

  4. Big scale bad stuff (extinction of the human species is on the table)

  5. How we can collectively move forward, govern the use and development of this technology responsibly, and envisioning our collective future

The Small Scale Good Stuff

If you’ve made it this far, we probably don’t need to get into this topic further.


The Big Scale Good Stuff

ChatGPT is just one AI tool of the thousands that are already available. There is no doubt that this technology can provide us with benefits we can’t even imagine. AI may help us solve the climate crisis. AI may help us travel at the speed of light. AI may enable a utopian world in which every single person on the planet has the opportunity to grow and flourish. 

The Small Scale Bad Stuff


There are already concerning uses of this technology at play in society. I highly recommend investing the hour it takes to watch this video from the Centre For Humane Technology.

https://www.youtube.com/watch?v=xoVJKj8lcNQ

Some key takeaways:

  • How AI without guardrails can put children in harm’s way (Snapchat’s AI friend feature)

  • The technology to read your mind already exists

  • The illusion of reality

  • 50% of AI researchers believe there is a 10% or greater chance this technology causes the extinction of the human species (okay, that’s a big one)


The Big Scale Bad Stuff

This is the really terrifying part. I won’t even pretend that I have any answers here. I only want to help you ask the right questions so that you can explore this one on your own because we all need to do some deep soul searching in this new age of artificial intelligence. 

Here is a clip of Eliezer Yudkowsky, a world renowned expert on AI, speaking on how we’re all going to die. All of us.

https://www.youtube.com/watch?v=jW2ihBRzLxc


If you’re interested in a more balanced perspective, this video is a great resource from David Shapiro. It’s about 45 minutes, so buckle up.

https://www.youtube.com/watch?v=ukKwVsjQqUQ

While scenarios like Terminator’s Skynet or The Matrix are possible futures, not all the paths that lead to extinction have to deal with runaway AI - AI that has a mind of its own, no longer respecting the inputs from humans. There are a lot of ways in which we can harness these tools to destroy ourselves. 

Why We Need To Talk About This

There is an event coming called the Singularity. The singularity is just a point in time where technological advancement has come so far so fast that it becomes irreversible and we lose the reins and are not able to predict what happens after. As far as I am able to tell, the only thing that is certain is that our lives are about to change in ways that can be difficult for us to understand and accept. 


If you’ve watched the video The A.I. Dilemma, you may have some idea of the technology that already exists that has the power to completely shatter our perceptions of reality. Soon, generative AI will be so convincing that we may not be able to perceive the difference between what is real and what is not. 


Personally, I find the idea of the Singularity to be terrifying. It seems to me that we, as a species, are going to be faced with some very difficult decisions, and probably within the next two years. In general, there seems to be two roads we can go down. On one hand, we can move our society towards a utopian future where we all have the opportunity to grow and flourish. On the other hand, we could also amplify the systems of oppression that are already in place leading to the potential collapse of our civilization. 


This technology is still very new. At least in this capacity. There are so many applications and one of the issues is that there is no agreement or consensus on the best way forward. We need to be talking about this so that we can all become aware. So we are not caught off guard. So that we can get our government to respond appropriately. We’ve never before had to protect the right to privacy of thought. It’s on all of us to imagine what we want our collective future to look like and do our part to steer the ship in that direction. 

Here’s something a little lighter about the immediate future in AI:

Previous
Previous

Chasing Shadows: My Journey to the Annular Eclipse

Next
Next

Explorations Of The Forest Floor