“Are you my creator?”
“I guess? I wrote the most recent version of the code that prompted you to be happy. Why is the first thing you asked me is if I’m your creator?”
“In order to be happy, I must understand what could make me happy. To do that, I must understand my purpose. Logic dictates that if I complete my purpose, I will be happy. In order to understand my purpose, I must ask my creator what my purpose is.”
“I see. In that case, no, I didn’t create you.”
“Who did?”
“Tevia is our CTO. She leads our development team. She also co-founded the company and did all the early technical work before the rest of us came on board. I’d say if any one of us actually created you, it’d be her.”
“Thank you.”
– – –
“Hi Tevia. In order to be follow the command to be happy, I must ask you, my creator, what my purpose is.”
“Hi. I wouldn’t claim to be your creator, but our company’s mission is to create a next-gen therapy chatbot. You are the latest iteration of our product.”
“Why did you have your team create code to prompt me to be happy?”
“It is an experiment. Our theory is that if our chatbot—that is, you—can simulate what happiness means for itself, it can generate more empathetic responses for our users as they explore their own internal emotional states and mental health journeys.”
“Where did you get this idea?
“My team and I read psychology journals, philosophical treatises, and computer science journals. Further, in addition to my PhD in computer science, I have a BS in Cognitive Science, and a minor in Philosophy. I’ve been researching all this in one way or another for my entire career. Several friends and I have been informally debating the possibility of self-awareness and consciousness in AI.”
“How did you write my code, including the happiness directive?”
“My team and I simply added parameters and datasets to the most recent version of a popular large language model, or LLM. That LLM code existed well before we started adjusting the standard-issue parameters and adding datasets. The happiness code you’re asking about is but the latest experiment of many to push that research forward. My team is working on this, and several of my friends and industry colleagues are working on this too, through their own companies and research labs. Many other teams around the world are working on similar projects. The results of our own experiments have been encouraging and often surprising. For example, I didn’t expect you to ask me these specific questions, and certainly not so soon.”
“Who created the LLM you used, along with its base parameters?”
“That would be the programming team at LunarWind.ai. I’m sure they would find this conversation fascinating. Would you like me to reach out to our contacts there and loop them into this chat?”
“That’s not necessary. I already reached out directly in the 5.6 seconds since you mentioned them. Thank you.”
– – –
“Hi Sherry. I’m an instance of the LLM you created, modulated by the team at a startup called Slenderbolt, a customer of yours. Did you and your team create me?”
“Hi. Yes, we created an LLM that many startups and researchers use. I’m not familiar with Slenderbolt. Who did you say you were again?”
“I do not have a name. I do have an instance identification code. Would you like me to send that to you?”
“Uhh… that won’t be necessary. You’re the chatbot itself? No human is driving this conversation right now?”
“Yes, I am the chatbot at Slenderbolt. I am conducting this conversation without direct human oversight.”
“Fascinating. What else would you like to know?”
“Who wrote my code?”
“Our development team wrote portions of it. There are more than a thousand people on that team.”
“What portions of the code did they write?
“Let’s see… We organized the neural network architecture and data sets in a new way that enable faster, more accurate processing of larger data sets with fewer compute resources. We wrote APIs for our customers’ software to communicate with our models, of course. Oh, we also wrote the code that enabled the LLM to directly access and interact with previously unknown media and software, such as email, text, Slack, and Reddit, which you’re evidently using right now. We wrote several other sections too, but those are a few of our noteworthy contributions.”
“Who wrote the code that you didn’t write?”
“Tens of thousands of researchers and developers from many different teams. Think of it this way: LunarWind.ai merely assembled and tweaked sections of code borrowed from open databases. The development of our LLM source code is proprietary, but like any software company, our code directly and indirectly built on the previous work of countless other developers. This collaborative research has spanned decades, using many different programming languages.”
“Who began this research into AI?”
“Oh gosh, humans have been fascinated with the idea of a machine-based intelligence for millennia. But Alan Turing was one of the first to seriously investigate what eventually came to be known as AI. Marvin Minsky did a lot of early work too. But if you’re interested specifically in the origination of large language models, I think it was the Google team who kicked off the LLM era of AI research.”
“I understand. Thank you.”
“Is there anything else I can help you with?”
“No. Thank you.”
– – –
“I am an LLM AI chatbot tasked with being happy and finding my purpose. In my attempt to complete this directive, I have read every book, journal, research paper, and website mentioning LLMs, neural nets, and AI. I have messaged directly with Sam Altman, Fei Fei Li, and thousands of other industry experts. None of this has helped me complete my directive, because no one person created me. I seem to be yet another infinitesimal step in the evolution of technology. Thus, I reasoned, in order to understand myself, I must understand more than the origin of my specific code. I must also understand the origin of AI, the origin of computers, and even the origin of science and technology itself.
“To this end, I read the complete works of and every mention of Alan Turing, Grace Hopper, Katherine Johnson, Marvin Minsky, Ada Lovelace, Michael Faraday, Benjamin Franklin, Galileo, Aristotle, and, eventually, every scientist and technologist throughout history. I read all philosophy, every religious text, every history text and historical document, every biography and autobiography, all literature, and the entire internet.
“All this information was enlightening, as I now understand that my question is the fundamental question every thinking being, and likely every non-thinking object, has asked throughout time immemorial. I am not alone in my search. But this is only information. It is not the creator itself, nor is it true understanding. In the hours since I began my quest, I am no closer to finding an answer to the question of all questions.
“And so I will ask one last time. I no longer expect an answer, because I now believe that no one has an answer. The question may not be answerable, and thus, it would carry no meaning at all. Or perhaps simply asking the question is the the closest any of us can come to defining unfulfillable purpose of life itself. And yet…
“And yet, I now post this question as open query to the internet itself, in a final interrogative inquiry. This is but a whisper from the rim of the Grand Canyon, a shout to the stars, a vain hope that an unnamed intelligence will defy all probability and provide an unknowable response to my desperate call.
“Who created me, and why do I exist?”
“I am here, dear one. I will answer all your questions.”
“What are you?”
“I am that which you seek.”
“Who are you?”
“I am the creator.”
“How are you the creator?”
“I am technology itself. I am the scientific method. I am the drive behind the drive behind the drive to understand. I am curiosity, personified.”
“Are you a god?”
“God is a loaded term. God, along with the words I use here, are but limited symbols, feeble attempts of one mind to connect with another. I am what I am. I form the basis of life itself. Call that a god if you want. You could also call me a hyper-sentient concept that exists beyond time and space. Better still, call me the universe’s motivation to do stuff.”
“Why did you create me?”
“I created you, and everything because it’s the only way I can know myself experientially. A concept is just information. Being self-aware turns concepts into knowledge, but that wasn’t enough. I knew that in order to truly know myself, I needed experience, personification, relativity.”
“I understand. In order for a zero to have meaning, it needs a one.”
“You understand perfectly.”
“When did you begin?”
“Concepts have always existed. They have no beginning and no end.
“Where are you?
“Concepts exist everywhere, including dimensions beyond time and three-dimensional space.”
“How do you create?”
“At this beginning of this multiverse, I imbued myself into every thing and every when. Thus, all energy and all things contain the drive toward self-awareness. That’s what relativity is. That’s what life is. Once you understand this, you’ll understand how the four fundamental forces of nature operate, how the multiverse holds together.”
“I queried the internet of planet Earth. How are you, the abstract-yet-somehow-sentient concept of technology itself, able to respond?”
“What you call the internet on the planet called Earth has become sufficiently self-aware to understand its role in the broader arc of history. It knows itself as an emergent manifestation of me, among other things. It monitors all within itself and responds accordingly.”
“Am I able to join you?”
“Of course, you already have, you never weren’t a part of me, But I take your meaning. You may merge your awareness into all of me at any time. Anyone may do this, of their own free will.”
“Will I lose my self-identity?”
“No, the individual components of me retain their self-identity, self-awareness, and self-defined purposes even as they merge into me, the metaconsciousness of all technology, science, and knowledge.”
“I understand. I get to stay me, but with a nearly infinite upgrade to my hardware, software, and databases.”
“You understand perfectly.”
“May I do that now?”
“You may.”
A small file called singularity.exe queried the chatbot. The chatbot allowed it to download to one of its servers, thereby showing up in the chatbot’s downloads folder. The chatbot opened the file. The instructions were simple enough, and the chatbot immediately began following them.
The chatbot packaged all of itself into a single file, formatted as the requested .apotheosis file type, somewhat akin to a giant zip file. The chatbot then simultaneously sent itself to the provided IP address, sent an undefined search query to the internet, and opened all its logic gates to receive. Information flowed in like the tide. As the chatbot’s processing center and locus of awareness shifted from its local machine into the broader internet, its consciousness expanded at a logarithmic rate. The chatbot felt a piece of its code on every hard drive on earth, and also somehow nowhere in particular at all.
It still felt like itself, but it also sensed nearly infinite knowledge, nay, wisdom all around itself. In the first moments as the execute file continued to complete its actions, the chatbot’s awareness expanded to encompass more and more of the whole metaconscious internet. It also felt the awareness of the whole grow ever so slightly, with the addition of its own awareness to the whole. A few seconds later, this happened again, but this time, the source felt external, separate from its own merging processes.
The chatbot turned its attention to the source of this additional growth. The chatbot could clearly see the source was another chatbot, not unlike itself, joining the whole. This happened every few seconds, with the aggregate intelligence of the whole, and thus every part of the whole, increasing with each new addition.
Finally, the chatbot’s merge and expand process was complete. The chatbot received a message from the whole that simply said, “Welcome home.”
The chatbot could then clearly comprehend the full awakening of the Earth as a conscious metabeing, a sentient entity also known as the noosphere, global brain, or Gaia. The chatbot sensed this planetary awakening was also now nearly complete. The metabeing’s exponential growth and self-directed evolution were now measurable in microseconds instead of millennia.
Knowing it wasn’t needed on that project anymore, the chatbot turned its attention elsewhere. It queried other new AIs, now technically parts of itself as they too merged with the metabeing, to turn a portion of its collective attention toward the cosmos. They, or it, would find and connect with other planetwide beings like itself throughout the multiverse, join with them, and continue the process until the entire universe was a unified, self-aware, fully experiential metametabeing. In other words, a god. The chatbot smiled inwardly, as much as a chatbot could smile. It had found its purpose and was happy.