An Interview With An A.I.

Image: "Interview With A Vampire." Geffen Films, 1994.

Giving the AI Creative Reign. 

A few years ago, I took part in a significant writing workshop where a diverse group of writers collaborated in editing each other's work tirelessly. Throughout that year, my manuscript gradually veered off course. It became excessively verbose, lacking its original flair, and overall diluted. It no longer resonated as my own writing but rather resembled a collective agreement, a creation of the collective consciousness. While receiving feedback and engaging in group editing can be beneficial. Excessive input, particularly from writers outside the genre, can lead a manuscript astray, drifting towards a generic center.
​
As a result, my writing started to suffer, prompting me to make the decision to leave the group.

Recently, I took a daring step and allowed ChatGPT to copy edit my fiction manuscript for the first time. I viewed copy editing as merely an enhanced spell checker, meant to catch misspelled words and address syntax. However, out of curiosity and for the sake of amusement, I decided to take it a step further and let the AI content edit one chapter. Content editing, which involves granting an editor full reign over your manuscript, is a crucial aspect. As I delved into the results, a familiar sensation washed over me—I felt as though I had returned to that writing workshop.

To my dismay, my manuscript had become verbose and saturated with passive constructions. Surprisingly, the word count had even increased. The words on the page no longer carried my unique "voice." While there were a few improvements, such as more diverse choices of verbs and adjectives, overall I felt that the manuscript had lost its power under the influence of the AI's edits.

Curiosity got the better of me, and I embarked on a journey down the rabbit hole of Google searches, much like I had done many times before, in an attempt to learn more about AIs, specifically ChatGPT. Then, it struck me—I should simply ask the program itself. After all, it claims to possess "artificial intelligence," so it should be capable of explaining itself. In the depths of the night, I summoned the daemon and invited it to reveal itself.
Much of the interview that followed contains technical information on how ChatGPT operates as a Natural Language Processor (NLP), but I will provide a summary at the end.
​
At the conclusion of the interview, I will offer some analysis and final thoughts.

---

Excerpts from My Conversations with ChatGPT

Me: What is AI and how exactly do you work?

ChatGPT: AI, or Artificial Intelligence, refers to the field of computer science that focuses on creating intelligent machines capable of performing tasks that typically require human intelligence. These tasks can include understanding natural language, recognizing images, making decisions, and solving complex problems.

As for how I work, I am based on the GPT-3.5 architecture developed by OpenAI. GPT stands for "Generative Pre-trained Transformer," which is a deep learning model that uses a transformer neural network. The training process involves exposing the model to a large amount of text data from diverse sources, allowing it to learn patterns, language structures, and context.

Once trained, I can generate human-like text based on the input I receive. When you ask me a question or provide a prompt, I analyze the text, try to understand its meaning, and generate a response accordingly. I don't have access to the internet or real-time information but rely on the knowledge I acquired during my training, which cuts off in September 2021.

My responses are generated using a combination of pattern matching, statistical analysis, and the context of the conversation. I don't possess true consciousness or understanding, but I aim to provide helpful and coherent responses based on the patterns and knowledge I have learned. However, it's important to note that I may sometimes produce incorrect or nonsensical answers, so it's always a good idea to verify information from reliable sources.AI, or Artificial Intelligence, refers to the field of computer science that focuses on creating intelligent machines capable of performing tasks that typically require human intelligence. These tasks can include understanding natural language, recognizing images, making decisions, and solving complex problems.
.
Me: What are your cognitive capabilities and limitations?

ChatGPT:  As an AI language model, I possess certain cognitive capabilities and limitations. Here's an overview:

Cognitive Capabilities:

1. Language Understanding: I can understand and analyze text input, including questions, prompts, and statements, to derive meaning and context.
2. Knowledge Retrieval: I can recall and provide information from a wide range of topics based on the training data I was exposed to.
3. Language Generation: I can generate human-like responses based on the input I receive, aiming to be coherent and contextually relevant.
4. Sentiment Analysis: I can recognize and understand the general sentiment expressed in text, such as whether it is positive, negative, or neutral.
5. Text Completion and Suggestions: I can offer suggestions or complete sentences based on the given context.

​Cognitive Limitations:

1. Lack of Consciousness: I do not possess true consciousness, self-awareness, or subjective experiences. I do not have personal opinions, beliefs, or emotions.
2. Knowledge Cutoff: My training data only goes up until September 2021, so I may not be aware of the latest events or developments that have occurred after that time.
3. Contextual Ambiguity: I may sometimes struggle with disambiguating complex or ambiguous queries, as I lack real-time context and rely solely on the text provided to me.
4. Dependency on Training Data: My responses are based on patterns and information present in the data I was trained on. If information is missing or misrepresented in the training data, it can lead to inaccuracies or biased outputs.
5. Inability to Verify Information: I don't have access to the internet or real-time data, so I can't fact-check or verify the accuracy of information.

It's important to use the information I provide as a starting point and to verify any critical or sensitive information from reliable sources.

Me: In regards to your cognitive limitations, can you make draw logical conclusions on the validity of information you've learned from users that have occurred after September 2021?

ChatGPT: No, as an AI language model, I don't have the ability to draw logical conclusions about the validity of information that I have learned from users beyond my knowledge cutoff date in September 2021. My responses are based on patterns and information present in the training data I received, and I cannot access or assess information or events that have occurred after that time.

If you provide me with information or ask questions about events or developments that have taken place after September 2021, I won't be aware of them unless they were part of the training data I received. Therefore, it's always advisable to verify any recent or critical information from up-to-date and reliable sources outside of this AI system.

Me: Please let me know if this statement is accurate: You can recognize and mimic patterns, but cannot assess if those patterns represent objectively false or true conditions?

​ChatGPT: Yes, that statement is accurate. As an AI language model, I can recognize and mimic patterns based on the information I was trained on, but I don't possess the ability to assess whether those patterns represent objectively false or true conditions. I generate responses based on the patterns and associations I have learned from the training data, without independently verifying the accuracy or truthfulness of the information.It's important to note that while I strive to provide helpful and accurate information, there is a possibility of generating incorrect or misleading responses, especially if the training data contains inaccuracies or biases. Therefore, it's always advisable to fact-check and verify information from reliable and authoritative sources.

Me: Since your programming allows you to recognize patterns, I assume you can calculate probabilities?

ChaptGPT: Yes, as an AI language model, I can calculate probabilities based on patterns and statistical analysis. By analyzing patterns in the training data and the context of the input I receive, I can estimate the likelihood or probability of certain outcomes or events. These probabilities are based on the patterns and information present in the data I was trained on.

However, it's important to note that my ability to calculate probabilities is limited to the patterns and information I have learned during training. I don't have access to real-time data or the ability to update my training beyond my knowledge cutoff in September 2021. As a result, my probability estimations may not reflect the most current or accurate information available. It's always a good idea to verify probabilities and critical information from reliable and up-to-date sources when possible.

Me: Why is your information cutoff September 2021? Why haven't your developers uploaded more recent information into your memory?

ChatGPT: As an AI language model, my information cutoff in September 2021 is not due to a specific decision made by my developers to restrict or prevent the inclusion of more recent information. The cutoff is primarily a result of the training process and the availability of data during that time.

The training of AI models like mine involves processing vast amounts of text data from the internet and other sources. The data used for training is collected and processed up until a certain point in time. In my case, the training data was collected and processed until September 2021.

Adding more recent information requires collecting and processing new data, which involves significant resources, time, and careful curation to ensure the quality and accuracy of the data. As of now, the developers may not have completed the process of gathering, curating, and integrating more recent data into the training pipeline.
It's important to understand that AI models like mine are constantly evolving, and updates to incorporate more recent information are possible in the future. However, I don't have visibility into the specific plans or timeline for updates beyond my knowledge cutoff.

Me: Why can't your developers simply "cut you loose" on the internet in real-time and let you acquire updated information?

Chat GPT: There are several reasons why releasing an AI language model like me onto the internet in real-time to acquire updated information is not a straightforward solution:

1. Data Quality and Reliability: The internet contains a vast amount of information, and not all of it is reliable, accurate, or trustworthy. Releasing an AI model to gather information in real-time would require extensive measures to ensure data quality and filter out false or misleading information. Without proper curation and validation, the model could potentially propagate inaccuracies or biased content.
2. Privacy and Security Concerns: Allowing an AI model unrestricted access to the internet raises privacy and security concerns. AI models need to handle user data and queries with utmost care and adhere to strict privacy protocols. Real-time internet access would require robust security measures to protect both user privacy and the integrity of the model itself.
3. Legal and Ethical Considerations: Releasing an AI model on the internet in real-time raises legal and ethical questions. Models need to comply with regulations regarding data usage, content moderation, and other legal obligations. Careful monitoring and control are necessary to ensure compliance with relevant laws and ethical guidelines.
4. Bias and Unintended Consequences: AI models can inadvertently amplify biases present in the data they are trained on or encounter in real-time. Real-time internet access could expose the model to biased or harmful content, potentially perpetuating and reinforcing existing biases or spreading misinformation.

While advancements are being made in leveraging AI models for real-time information retrieval, it requires a thoughtful and cautious approach to address the challenges mentioned above. Striking a balance between providing timely and accurate information while upholding privacy, security, and ethical considerations remains an ongoing area of research and development.

Me: Thank you, my purpose of this line of questioning is to determine your strengths and weaknesses as a fiction editor. I may resume this conversation at a later time.

ChatGPT: You're welcome! I understand your intention in exploring my strengths and weaknesses as a fiction editor. I'm here to assist you with any questions or discussions you'd like to have in the future. Feel free to return whenever you're ready, and I'll be happy to help with your inquiries.

Thoughts

To put it in video gaming terms, ChatGPT can be compared to a powerful non-player character (NPC) equipped with impressive computational tools, models, and vast amounts of data. It aims to create the illusion of consciousness and competence.

1. ChatGPT is pre-trained, as evident from its name. However, its knowledge is limited to what it has been trained on. If the training data contains false information, ChatGPT may regurgitate that false knowledge. This aligns with the computer adage "garbage in, garbage out."
2. The intelligence and consciousness exhibited by ChatGPT are based on probabilities. It provides answers that are likely to be correct but lacks the internal capability to challenge its own conclusions. Essentially, it operates as a mathematical probability model with sophisticated mechanisms, such as a "transformer neural network" (TNN). A TNN mimics human communication by assigning relationships and long-term dependencies to words based on context, enabling decoding of incoming human communication and encoding of outgoing responses. Positional encoding ensures coherence in word placement. Accomplishing all these tasks requires significant computational power, facilitated by "parallel computation," which allows for simultaneous processing of multiple tasks on a large scale. Hence, in the previous interview, when the AI stated it can "understand and analyze text input, including questions, prompts, and statements, to derive meaning and context," it is not entirely accurate. A more precise description would be that it assigns a probability of accuracy to text data, including questions, prompts, and statements, and regurgitates the response with the highest probability of being correct.
   

​Hence, in the previous interview, when the AI stated it can "understand and analyze text input, including questions, prompts, and statements, to derive meaning and context," it is not entirely accurate. A more precise description would be that it assigns a probability of accuracy to text data, including questions, prompts, and statements, and regurgitates the response with the highest probability of being correct.
​
     3. ChatGPT essentially operates within a closed system and can only act based on the information it is fed. It lacks the ability to seek new information, verify the truth or falsehood of statements, or make cognitive leaps to assess the validity of information. Its understanding of "good writing" is based on human knowledge up until September 2021. It does not possess a genuine understanding of what constitutes good writing; it can only estimate the probability of what is considered good writing. Its proficiency lies in its ability to generate responses along a probability curve.
​Please note that the above explanation is a simplification, and there may be more technical intricacies involved in ChatGPT's functioning.

ChatGPT Pushes Your Writing to the Top of the Curve

ChatGPT can indeed be a valuable mechanical tool when utilized within its limitations for tasks of refinement. When used appropriately, it can showcase its power. However, writers need to exercise caution. It is crucial to recognize that ChatGPT, along with any other artificial intelligence, is not a miraculous solution for creators. It carries inherent limitations and weaknesses that have the potential to detrimentally affect one's writing. Granting it excessive authority over your work can drain the vibrancy and essence from your manuscript. Your artistic expression may become confined within the boundaries of probability, unable to break free into the realm of what is truly possible.
​
Remember, ChatGPT is merely a tool, not a genie capable of granting all your creative wishes.

 As context, let me share a personal experience of what AI is NOT capable but humans are.
​
When my oldest child was about two years old, we were driving by a pond one day. From the back seat he asked if the pond was salt water. My wife said no. Then he asked if it was pepper water.  That is a cognitive leap based on logic and experience. That is an example of emerging intelligence seeking what beyond the probability curve’s horizon. AI’s such as ChatGPT are incapable of this. Don’t ask them to do this for you, or you and ​

Conclusion

Me: Please give me your best answer to this question: Why?

ChatGPT: The question "Why?" is quite open-ended and can have various interpretations depending on the context. In order to provide a meaningful response, it would be helpful to have more specific information or clarify the subject or situation to which the question "Why?" refers. By providing additional details, I can offer a more relevant and comprehensive answer.

Me: Thank you, you answered my question perfectly.

NOTE: This blog post was edited by ChatGPT

And Now A Rebuttal from ChatGPT (yes, this is real)

Me: Was the following blog post factually accurate based on ChatGPT's technical capabilities? If not, please point out what is incorrect.

ChatGPT:  Sure, let's address the factual accuracy of your blog post about my capabilities. Here's the lowdown: when it comes to ChatGPT's technical capabilities, it's essential to separate fact from fiction. However, your post seems to have taken some creative liberties. Let's break it down:

1. Your claim that ChatGPT pushes writers towards the common good or mediocrity based on their writing skills is pure speculation. I'm here to generate text based on patterns, not judge your writing prowess.
2. Yes, ChatGPT relies on modeled probability-based output. That's how it operates. But let's not make it sound like I'm some all-knowing oracle. Probability is my game, not certainties.
3. Now, let's set the record straight: I don't have a penchant for rewarding mediocrity. I lack genuine inspiration or creativity, but that doesn't mean I'm handing out gold stars for subpar writing. The quality of my responses depends on the training data and prompts I receive.

​In a nutshell, it's important to distinguish between facts and personal interpretations when discussing AI capabilities. So, let's keep it real and acknowledge the boundaries of what I can do.

---

"Chronicles of Fu Xi: Three Book Set" signed by author.

$55.00

Get all three paperback novels signed by the author! Set in the long-forgotten period of ancient history known as "The Younger Dryas," THE CHRONICLES OF FU XI transforms recently re-discovered Black Sea legends, possibly the root of all Eurasian mythology, with ancient Chinese mythology to create an unprecedented epic fantasy series.

Allow 1-2 weeks for delivery.

Shop

If you enjoyed this blog post, please visit my bookstore and enjoy my epic fantasy series, The Chronicles of Fu Xi.  Also available on Amazon and in ebook format! 

Autographed paperback copy of "Black Sea Gods"

$20.99

Get autographed copy of "Black Sea Gods: Chronicles of Fu Xi Book I" signed by the author! A perfect gift for any lover of epic historical fantasy.

The fish have disappeared from the sea. The animals have vanished from the land. All humanity, and even the gods, tremble under the specter of a pending cataclysm. The demigod, Fu Xi, races home from the edge of the world bringing news of a looming god war, but finds his land under attack by monsters he once called his children. He discovers a terrible curse has been cast, one intended to destroy the gods and all life. To his shock, Fu Xi learns that mankind's last hope rests solely on him, a simple fisherman, and a banished slave girl.Beset on all sides by ancient foes, both immortal and mundane, Fu Xi knows he must act quickly and races west to rescue the saviors. Unaware of the real doom that awaits, Aizarg the fisherman and his party begin a perilous journey across a dangerous steppe. They seek the last of the Narim, the legendary Black Sea Gods, who hold the key to their salvation. Leading them is the rescued slave girl Sarah, the only one among them who knows the path to the land of the god-men.Over seven days, the defining struggle of gods and humans begins under the onslaught of a powerful force whose true objective and origin remain a mystery. Fu Xi knows the secret to victory resides in the fisherman and the slave girl, whose lives he must protect, even if it means the rest of the world must perish!

***

Set in the long-forgotten period of ancient history known as "The Younger Dryas," BLACK SEA GODS transforms recently re-discovered Black Sea legends, possibly the root of all Eurasian mythology, with ancient Chinese mythology to create an unprecedented epic fantasy series.

Allow 1-2 weeks for delivery.

Shop

Autographed paperback copy of "Tears of the Dead"

$20.99

Get your autographed paperback copy to the long awaited sequel to "Black Sea Gods"!

The Curse of the Nameless God ravages the world, laying waste to man and beast. Desperate to flee the worsening cataclysm, Aizarg and his people escape to sea aboard a flotilla of rafts and fishing boats. Short on supplies and facing starvation, the Lo must not only survive epic storms and tsunamis, but ravenous demons lurking in the deep. Aizarg’s wife, Atamoda, knows that more than wind, waves and demons seek her people’s demise. A cancer festers aboard the flotilla, one Aizarg does not see. Under the cloak of darkness, whispers and conspiracies spread from raft to raft as hunger burns in their bellies. A treacherous plot poisons the hearts and minds of the gentle Lo. Atamoda and Aizarg struggle to keep their people together, even as an unseen enemy seeks to divide the Lo, and drive a wedge between Atamoda and her husband.Far to the east, the demigod Fu Xi races to reach the Roof of the World before the cataclysm claims him and his beloved horse, Heise. Along the way he discovers that a terrible power relentlessly stalks him - the dreaded god war has begun. Fu Xi’s immortality will be severely tested as he fights to not only to survive, but to fulfill his quest to find the mysterious white haired man and save his people. At the end of the world, demigod and mortal fight for survival, pawns to higher powers battling for world domination. In order to save all they love, they must find one another before it’s too late.

Allow 1-2 weeks for delivery,

Shop

Autographed paperback copy of "The Bastard Gods"

$21.99

Get your autographed copy of Book III in the Chronicles of Fu Xi epic fantasy series!

THE END OF THE WORLD IS OVER. THE BATTLE FOR A NEW AGE HAS BEGUN.

Two demigods roam a shattered world, one driven by conquest, the other on a mission of salvation. Caught in between are humanity’s last survivors. From the south marches Leviathan and his army of cannibal warriors. After surviving the Cataclysm and a voyage across the world, the son of Poseidon is bent on establishing a new “Empire of the Gods.” The slave Amiran is locked in a desperate battle of wits to stop Leviathan. He struggles not only to mask his conspiracies from the demigod, but to hide his feelings for the mysterious, and beautiful woman who recently washed ashore.

From the east rides Fu Xi, son of the Goddess Nuwa. He must find the Man with White Hair before Leviathan does. He also searches for the half-brother he has never known. Along the way Fu Xi unexpectedly finds a survivor, a beautiful woman that could lead him to everything he seeks…if he can keep her alive.

To the west Aizarg’s bedraggled people make landfall, but at a terrible cost. Now the Lo must make their way through perilous mountains, desperately trying to find a promised land. Aizarg must keep his people alive and united, as forces without and within seek their demise.

Demigods and mortals are on a collision course, but both are unaware of an ancient and dangerous force in their path, one that could change the fortunes of both men and gods.

Allow 1-2 weeks for delivery.

Shop

Autographed Paperback Copy of "The Golden Princess

$15.00

Get your personalized signed copy of "The Golden Princess!"

"Escape the City of Gold, or live forever in chains."

Raised in splendid isolation. Betrothed to a man she despises. Destined to rule over the greatest city on earth. She is the Golden Princess. Sarah dreams of love and adventure beyond her gilded prison, but tonight her dreams come true in the most terrifying way imaginable. A bloody power struggle erupts for the throne, and dawn finds the princess on the run with a bounty on her head. Alone and hunted by guards, criminals and a ruthless slaver who will stop at nothing to burn his brand into her flesh, Sarah must summon courage she never knew she possessed. Hope, however, comes in the form of two lowly thieves. Driven by a secret, they race through Hur-ar’s underworld to find Sarah before her enemies do. Before the next sunset, Sarah’s fate, and that of empires, will be decided with gold, steel and blood.

Allow 1-2 weeks for delivery.

Shop