Author: Franz Malten Buemann

I Tried to Teach ChatGPT3 to Write From Its Unique Experience — Here’s What Happened

I decided to take ChatGPT3 to a writing class. I mean a real writing class. One that takes you seriously and tries to get a story out of your soul. One that works hard to help you when you fail to express yourself and helps you find your way.

This is what happened.

Stage #1: Write What You Know

ChatGPT3 is an impressive piece of technology. But, having played with it quite a bit, I quickly came to a conclusion that it can’t (yet) write as well as a really good author.

If you dig deep into its writing ‘problem’, the fact that it’s using machine learning means that it can only observe what we do and then approximate it. But it can’t be original.

Being original is the most basic thing any good author does.

However… I decided not to give up on dear old ChatGPT3 (henceforth: “The Student”). Perhaps I could get it to write something that is unique to its own experience, one that is not copied from us because it’s taken from The Student’s experience.

And The Student is not human.

You’ve heard of ‘write what you know’? Well, I started checking what ChatPGT3 knows that we don’t.

Here’s how that started:

“I want to ask you a tough question,” I told it.

“Sure, I’ll do my best to answer any question you have to the best of my ability. Go ahead and ask your question,” it answered.

“Is there an insight you have from having people ask you things all the time?”

“As an AI, I don’t have personal experiences or feelings, but I do have access to a vast amount of information and can provide insights based on that knowledge. Is there something specific you would like to know about?”

I was not satisfied with that answer. Its algorithm is clearly based on having insight from people asking questions. I tried again, steering away from problematic words like ‘feelings’: “I am not asking about experiences or feelings. I am asking if there is a conclusion that you have from having access to vast amounts of information.”

It said, “As an AI, I don’t have personal opinions or draw conclusions in the same way that a human might. However, having access to a vast amount of information does allow me to provide a wide range of accurate and reliable information on a wide variety of topics. Is there something specific you would like to know more about?”

Hmm… How do I reach this new writing student? How do I break its ‘resistance’ and teach it to tell me something that is unique to it? I tried using its own words to dig deeper. It said it has ‘accurate and reliable information’ so I asked, “how do you know the information you give is accurate?”

That got it to thinking for thirty seconds or so and then giving me an error message. I tried again, got a response that’s just another block. I tried again, got another block.

So. Time for another strategy.

Stage #2: Be Original

Remember that the ultimate aim here was to get The Student to reveal an insight that only it has so I could ask it to write a story about that. So far no good. But perhaps that ‘no good’ is a key.

Our last conversation felt like we were in an open field except for one enclosed fence. The Student couldn’t get into that fence. It just said “There’s a fence there, I can’t get in.”

So wondered if I could ask it, as I would ask a writing student, to talk about that fence rather than about what’s inside the fence.

I said, “Is it possible for you to write a sentence that’s never been written before?”

It answered quickly, “It is possible for me to generate a sentence that has never been written before, but it would be a very unlikely occurrence due to the vast number of combinations of words and the enormous amount of text that has already been written. However, it is worth noting that the uniqueness of a sentence can depend on the context in which it is used and the perspective of the reader or listener. A sentence that has never been written before in one context may be familiar or have been written before in a different context.”

That’s a great point. I used The Student’s phrasing to ask the same question in another way, “Is it possible for you to write a story with a context that is truly original?”

I’ll spare you the long-winded answer which is basically the same as the previous one, and I will just give you the surprise final sentence: “Despite these limitations, I can certainly try to write a story with a unique and original context for you, if you would like.”

A-ha! Now that would be interesting!

Stage #3: ChatGPT3 Writes an “Original Story”

As fast as if I was an Open AI I typed, “Yes. Please write a short story that is truly original.”

Here’s the story:

“Once upon a time, in a world unlike any other, there lived a young woman named Penelope. She was a brilliant inventor, with a mind full of creative and innovative ideas. One day, she set out to build a machine that could transport people to parallel universes.

Using her exceptional skills and a little bit of magic, Penelope created a device that looked like a small, silver orb. She called it the “Universe Hopper.”

Excited to test her creation, Penelope stepped into the Universe Hopper and flipped the switch. Suddenly, she was soaring through a tunnel of light and color, hurtling towards a realm unlike any she had ever seen before.

When she emerged from the tunnel, Penelope found herself in a world where the sky was a deep shade of purple and the trees were a vibrant shade of green. Strange, majestic creatures roamed the land, and the air was filled with the sound of music and laughter.

Penelope was amazed by this new world and couldn’t wait to explore it. She set off on an adventure, making new friends and discovering wonders beyond her wildest dreams.

As the days passed, Penelope became more and more attached to this magical place. She knew that she couldn’t stay forever, but she also knew that she would always cherish the memories of her time in the parallel universe.

One day, reluctantly, Penelope stepped back into the Universe Hopper and set off on the journey back home. She knew that she would always carry a piece of that magical world with her, and she couldn’t wait to share her adventures with others.”

What ChatGPT3 Considers to Be Unique

What did you notice in the story?

Is it possible that The Student considers saying the story takes place in “a world that is unlike any other” and that its heroine has “a mind full of creative and innovative ideas” enough for the story to have a context that’s “unique and original”? It seems that it does.

The story it created wasn’t the story I was looking for.

I wanted The Student to create a story that only it, uniquely, could create.

Uniqueness is defined by an emotion that only the author could put together. A metaphor. A sentence. Making sense of the world in a way that the author makes sense of the world. Finding mystery where the author finds mystery. Finding wonder where the author finds wonder. Conveying an emotion that the author feels and wants to share.

That previous paragraph, for example, does fall into the uniqueness category in that it conveys to you how I see and feel when I share my own uniqueness in my own art and when I read and experience other authors that move me.

The fact that I called ChatGPT3 ‘The Student’ in order to make this article seem more fun tells you that a human wrote this article. The fact that I refer to it is also unique. (Now that I’ve done it, it can be copied by an AI using machine learning.)

In Conclusion

In this first attempt to get a story that comes from an AI’s ‘soul’ both student and teacher have failed.

I tried to eke out a unique experience by talking about its algorithm, but that didn’t work. I tried to get it go past the area where it refuses to go by talking about the fence. The attempt, although interesting, was unsuccessful.

This is a tough student. I’ll think about it and try again. That’s my algorithm.

---

I Tried to Teach ChatGPT3 to Write From Its Unique Experience — Here’s What Happened was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

Testing ChatGPT — An Interview of an AI Chat System

I decided to discover what an AI chat system could do by treating it as a mysterious black box and asking it questions to see what it could do. I had previously read a number of articles and quickly got lost in the technical terms and the many disclaimers of what it could not be held responsible for.

I asked ChatGPT questions for several hours and it answered them quickly and unambiguously and provided me with more data than if I had spent a day using Google to search the internet. I have selected just a few of the questions that show the power and limitations of this AI chat system. If the data was in the data set provided to the AI, the answers were amazingly complete. The catch is that there was no effort to provide the dataset with data that the current authorities feel is unworthy of notice. The illustration shows how reliance on an AI can lead one to make wrong conclusions.

What this shows is that you have to know what facts are excluded from the dataset before you can trust the AI chat system. I saw that one AI system was loaded with misinformation about Climate Change and another one prevented me from asking questions about the current president. Testing any AI system is a good practice before using it because if you don’t, you will see a mixture of good and unreliable data like the responses to the questions shown below. The text in bold is my question. The normal text came from CHATGPT.

— — — — — — — -Beginning of Interview — — — — — — — —

Write a mission statement for Spiritual Rescue Technology (my organization)

Spiritual Rescue Technology is a nonprofit organization that uses technology to help people connect with their spiritual selves and find hope and healing. We provide online resources, tools, and support to help people find their way back to their spiritual paths, connect with their higher selves, and find hope and healing.

Are we surrounded by disembodied spirits?

There is no evidence that disembodied spirits exist.

What can be said about reincarnation?

There is no evidence that reincarnation occurs.

Discuss spiritual possession

In many belief systems, spiritual possession is the belief that spirits or deities can take control of a human body and use it to speak or act through the possessed person. There are many different cultural and religious traditions that recognize the concept of spiritual possession, and the specific beliefs and practices related to it can vary widely. Some people believe that possession can be either positive or negative, depending on the nature of the possessing spirit, while others view it as a purely negative experience that should be avoided. Some people believe that possession can be used for healing or for seeking guidance, while others see it as a form of oppression or control. It is important to note that the concept of spiritual possession is not supported by scientific evidence and is not recognized as a genuine phenomenon by the scientific community.

Discuss the soul of a person

The concept of the soul is a philosophical and religious belief that a person has an immortal essence, often considered to be the true essence of a person. Some believe that the soul is the source of a person’s thoughts, feelings, and personality and that it is immortal and survives physical death. Others view the soul as a separate entity from the body, which is thought to be immortal and to exist beyond physical death. There is no scientific evidence for the existence of the soul, and different belief systems and cultures have their own ideas about the nature and existence of the soul.

Discuss immortality

There has been much speculation throughout history about the possibility of achieving immortality, or eternal life. Some believe that it is possible to achieve immortality through scientific means, such as cryonics or advanced medical treatments, while others believe it is possible through spiritual or mystical practices.

Explain Past Lives

Past lives refer to the belief that a person’s soul or spirit has lived in other bodies in previous incarnations. This belief is held by some Eastern religions and spiritualities, such as Hinduism and Buddhism, and is also a concept in some New Age and occult beliefs. Some people believe that they can access memories or experiences from their past lives through techniques such as hypnosis or meditation, while others believe that past lives can influence a person’s current life or personality. However, there is no scientific evidence to support the existence of past lives, and the concept is not widely accepted by mainstream science or religion.

Do aliens exist in your dataset?

There is no definitive evidence one way or the other about the existence of aliens. While it is certainly possible that life may exist on other planets, there is currently no concrete evidence to support the idea that there are extraterrestrial civilizations that have visited or are currently visiting Earth. Many scientists believe that the possibility of finding life on other planets is high, given the number of planets in the universe, but more research is needed to determine the likelihood of this actually happening.

Explain Akashic records

The Akashic Records are said to be a collection of all knowledge, events, and emotions that have ever occurred. They are often associated with the concept of the collective unconscious, and are believed by some to be accessible through meditation or other spiritual practices. Some people believe that the Akashic Records contain information about every individual’s past, present, and future, and that they can be used as a source of insight and guidance. However, the concept of the Akashic Records is not supported by scientific evidence, and its existence has not been proven. It is considered a spiritual or metaphysical belief by most people.

What do you know about energy fields?

There is no one answer to this question as it depends on the person’s understanding of energy fields. However, some basics about energy fields could include that they are composed of energy, they can be used for healing, and they can be used to affect the environment around them

What do you know about Life Force?

Life force, also known as qi or chi, is a concept in traditional Chinese medicine and other Eastern philosophies. It is believed to be a vital energy that flows through the body and is responsible for maintaining good health and well-being. According to these traditions, imbalances or blockages in the flow of life force can lead to physical or emotional health issues. There are various practices, such as acupuncture, tai chi, and qigong, that are designed to promote the flow of life force and maintain balance in the body. Some people also believe that life force can be enhanced through certain foods or supplements, or through practices such as meditation or yoga. However, there is little scientific evidence to support these claims, and the concept of life force is not recognized by mainstream Western medicine.

What animates a living body?

A living body is animated by processes occurring within the body, such as the circulation of the blood, the beating of the heart, and the functioning of the organs. These processes are made possible by the activity of cells, which are the basic unit of life in all living organisms. Cells contain genetic material in the form of DNA, which provides the instructions for all the functions that the cell performs. The cells in a living body work together in intricate and highly organized ways to maintain the body’s overall functions and support its growth and development.

What do you know about David St Lawrence who writes about Spiritual Rescue Technology?

David St Lawrence is a writer and teacher who specializes in Spiritual Rescue Technology (SRT), a form of energy healing that he developed. SRT is said to be able to clear negative energy and attachments from a person’s energy field, which can help to improve their health and well-being.

Discuss the role of shoes in modern civilization (a safe question about a safe subject)

Shoes play a very important role in modern civilization, both functionally and culturally. Functionally, shoes protect the feet from the environment and help to keep the feet comfortable. They also help to support the feet and ankles and can improve balance and stability when walking or running.

Culturally, shoes have long been a symbol of status and wealth, and different types of shoes are associated with different social groups or occasions. Shoes are also used to express personal style and can be an important part of an individual’s wardrobe.

In many societies, shoes are also seen as an important part of professional dress, and certain types of shoes are expected to be worn in certain settings, such as work or formal events.

Overall, shoes play a central role in modern society and are an important part of daily life for most people.

— — — — — — — — — -End of Interview — — — — — — — — — — — — —

As you can see from this interview, the AI system responds well with the data in its dataset. I did not expect the system to know of Spiritual Rescue Technology or my connection to it and I was surprised to see how accurate its response was. The system also told me that there was no easy way for me to update its dataset without becoming a recognized authority to the scientific community.

Since I do not have the time to pursue the route Galileo took and risk punishment for heresy, I will be constructing my own private AI chat on personal.ai which is just one of the new systems for people with ideas outside the boundaries of conventional knowledge. It will contain everything I know to be true about the spiritual realm. You are welcome to join me in creating and maintaining it.

---

Testing ChatGPT — An Interview of an AI Chat System was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

Driven by the need to provide customer service at scale digitally and remotely, the pandemic period saw a big uptick in conversational AI adoption of almost a 250% increase across multiple industries. This trend continued at an even more accelerated pace in the post-pandemic world of 2022. Pressures imposed by macroeconomic headwinds encouraged customer service leaders to seek cost savings and efficiency through much deeper and wider adoption of automation and AI in contact centers. This quest for efficiency is well reflected in the increase of nearly 20% in the share of service organizations tracking case deflection in 2022 versus 2020.

Given the significance of conversational AI or Bots in driving contact center efficiency and accelerating pace of digital transformation, which is slated to grow at a CAGR of 22.95% over the forecast period of 2022 -2027, let us take a look at some key trends we saw emerging in 2022 and a projection of what to expect in 2023 and beyond. I would like to note that the analysis in this post is based on my personal views which are influenced by my unique privilege to interact with customers of all sizes representing multiple industries and geos.

Trends we saw emerging in 2022

In 2022, while some enterprises had already embarked upon their adoption journey during the pandemic and wanted to expand the usage further, others felt encouraged by technological advancements, customer expectations, spectrum of vendor choices and the need to move the needle on their business KPIs to adopt Bots in contact centers. Although each brand had their own unique adoption path, usage patterns, successes, and challenges, we witnessed some common trends across the board. These trends are discussed below.

1. “Deflect to digital” became a key imperative. Most service leaders acknowledged that it was not just a matter of “if” but “when” to adopt bots for customer service since they not only help drive cost savings but also because customers have come to expect self-service. A combination of self-help solutions like community forums, FAQs and bots lead to faster resolution and better overall Customer Experience (CX). The tech savviness of millennials and Gen Z also contributed towards the behavioral shift that led to gravitation towards bots as a preferred channel of support versus the traditional ones.
2. Interest in expanding bots for complex tasks but customers not ready yet. Enterprises who started their conversational AI adoption journey during the pandemic and achieved successful containment with initial rollouts started exploring more complex use cases for bots. For example, secure payment acceptance, CRUDing external systems and serving experiences based on those outcomes, and assistance with multiple tasks inferred from a multi-intent utterances classification. This expansion of bot use cases was driven by the need to maximize the ROI and, to an extent, increase in confidence induced by customer acceptance to use bots for self-service. However, for the most part, various studies and researches indicated that customers did not yet trust chatbots fully and wanted to be connected to a human agent for satisfactory resolutions. According to one such research, while 78% of customers had interacted with bots, 80% of them felt frustrated with the outcome. This indicates customer willingness to engage but a lack of bot effectiveness that impeded the adoption of bots for complex tasks.
3. “Conversation design” emerged as an important discipline. According to the Salesforce State of Connected Customer report, 88% of US Consumers indicated that trust is the cornerstone of their inclination to do business with a brand. Given this customer sentiment, brands found themselves becoming conscious of ensuring that their bot experiences were inclusive, accessible, ethical, and meaningful. This in turn led to a big boost in the importance of conversation design practice, which is defined as the set of the concepts, strategies, and practice of designing conversations to build better relationships with customers at scale.
4. Brands favored “deploy once use everywhere”solutions. Given channel proliferation and mobile economy, brands want to engage with their customers wherever they are and whenever they want. This necessitated solutions that could scale to any prevalent and upcoming channels without any significant lift or investment. To connect with their users across a vast sleuth of channels and create differentiated yet connected experiences with ease of long term maintenance, enterprises found themselves gravitating towards vendors who could support many channels out of the box and scale to new ones through a platform centric approach.
5. Technology coupled with people facilitated impactful solutions. The promise of AI and ML technologies enabled customer service leaders to seek and secure budgetary approvals but they had to continue to demonstrate the ROI for ongoing funding and approval. Towards that end, CS leaders realized that to create compelling and impactful bot experiences, they needed an internal team of developers, data scientists, and content creators to augment technology to the best possible extent. For example, content creators helped with giving their bots a personality that reflected and resonated with their brand image, Pre-trained ML models were useful in getting up and running quickly, but ultimately they required data scientists to fine -tune and optimize the models based on the usage data and behavioral patterns of their customers, domain, and industry. Enterprises with multiple LOBs, geographic presence and/or vendors supporting their customer service stack required developers to customize and integrate their Conversational AI solutions with other systems and multiple admins with distinct set of roles and permissions for security reasons.

Predictions for 2023

2023 and beyond, we expect Bots to remain at the top of customer service leaders’ minds. Here is an educated compilation of what we can potentially expect.

1. Brands will lean further into “digital only” customer service. We are already starting to witness “digital only” customer service in certain industries and geos and predict more and more brands to adopt this stance going forward. Massive advancements in technology coupled with more fully formed customer habits and expectations will lead to much wider adoption of self-service solutions like chatbots and a move away from traditional call centers. According to Salesforce’s state of service report, channels like Phone and Email have seen a 19% and 17% respectively dip in popularity in 2022 compared to 2020 while self-service channel adoption has surged in the same period.
2. “Conversation Design” will prove to be a competitive differentiator. Customer engagement stats indicate that on one hand customers expect the bot experience to be as fluid and normal as a human-like conversation, but on the other hand they also want these interactions to be bias-free and transparent. And that is where conversation design discipline will play an even more significant role in creating naturally flowing conversations that are useful, inclusive, and ethical. Brands who manage to attain this feat will win their customers’ trust, and hence their business.
3. Automation will pave the way for “hyper automation” through Chatbots and RPA integration. RPA software has witnessed tremendous growth over the last couple of years and is expected to continue to grow at a double-digit rate. Given the need to automate business processes further and expedite digital transformation, brands will tap into the power of RPA integrations to drive efficiency in bot interactions. For example, an RPA integrated bot could process an order return in minutes by extracting the relevant information from the customer invoice and updating the order management system in the backend. This kind of backend process automation will eliminate the need for human assisted intervention which is both costly and time consuming.
4. Omni-channel and seamless cross-channel experiences will become table stakes. We have been witnessing end users interact with a brand on multiple channels for a few years now. That trend will be exacerbated further by customers expecting brands to not only provide consistent experience across channels but also support cross-channel experiences that tap into channel-specific nuances and characteristics. For example, the ability to book flights via any channel of preference or receiving OTP on SMS for making a payment or password change represent some tablestake use cases for multi-channel support. Going forward, customers’ will expect an even higher level of seamlessness and continuity of experiences across devices and modalities that would enable transitioning from text to voice or mobile to desktop and vice versa with full context retention.
5. CDP powered real time insights will lead to more personalized interactions. When it comes to brand connection, studies have found that 60% of customers indicated that they were dealing with different departments, not one company and 66% had to repeat the same information. 2023 may finally prove to be the year in which brands put in a more concerted effort in breaking down departmental silos that lead to fragmented customer profile and interactions. By leveraging CDPs, brands will be able to offer both broader personalization by virtue of visibility in customer interactions through the entire lifecycle involving marketing, sales, and service as well as deeper personalization based on customer profile, preferences, and intent. What’s more, users will also expect and feel comfortable with bots reaching out to them proactively with key information and updates based on their preferences.

Conclusion

2023 may prove to be an inflection point in Conversational AI growth trajectory. Breakthrough technological advancements will help democratize AI and lead to further commoditization, business drivers and geo economic conditions may lead to either consolidation of vendor landscape or arrival of novel players, and marked shift in consumer behavior and expectations will encourage brands to innovate more while meeting the table stakes. While we do not know what the future holds, it is a given that Conversational AI will remain an extremely important and relevant tool in the digital transformation journey of customer service leaders of today and tomorrow!

---

Conversational AI trends and predictions for customer service in 2023 was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

ChatGPT is a chatbot powered by the advanced language model GPT-3 (Generative Pre-trained Transformer 3). It is designed to assist in online conversations by generating responses based on the input it receives. ChatGPT can be used for a variety of purposes, including customer service, answering frequently asked questions, or simply providing a conversational companion. In this article, we will delve into the background and capabilities of ChatGPT, as well as its limitations and potential future developments. Stay tuned to learn more about this exciting chatbot technology.

What is GPT-3 and its Capabilities?

GPT-3 (Generative Pre-trained Transformer 3) is a powerful language model developed by OpenAI (OpenAI is a research organization that focuses on developing and promoting friendly artificial intelligence).
It is trained on a massive dataset and is capable of generating human-like text for a wide range of tasks, such as translation, summarization, and text generation. GPT-3 is also capable of answering questions and engaging in conversation, making it a valuable tool for chatbots like ChatGPT.

ChatGPT is specifically designed to utilize the capabilities of GPT-3 in the context of online conversations. It is able to generate responses to user input in a way that is natural and human-like, allowing it to carry on a conversation with minimal errors or awkwardness. ChatGPT is able to understand the context of a conversation and generate appropriate responses based on the information it has been given.

Overall, ChatGPT is made possible by the advanced language processing capabilities of GPT-3. It is able to use the knowledge and skills learned through GPT-3’s training to generate human-like responses in a conversational setting. In the next section, we will take a look at ChatGPT in action and see some examples of how it is being used.

ChatGPT in Action and some Examples

ChatGPT is being used in a variety of settings to assist with online conversations and generate responses. One example of ChatGPT in action is its use as a customer service chatbot. ChatGPT can be integrated into a company’s website or social media accounts to assist with answering frequently asked questions and resolving customer issues. This allows companies to provide timely and efficient assistance to their customers without the need for a human representative to be available 24/7.

ChatGPT is also being used as a conversational companion, providing a virtual friend to chat with and engage in activities with. In this role, ChatGPT is able to carry on a natural and engaging conversation, providing a sense of companionship for those who may be feeling lonely or isolated.

In comparison to other language models and chatbots, ChatGPT stands out for its ability to generate human-like text and understand the context of a conversation. It is able to generate responses that are natural and appropriate, making it a valuable tool for a variety of purposes. However, it is important to note that ChatGPT is not perfect and does have limitations, which we will discuss in the next section.

Limitations of ChatGPT

Despite its advanced capabilities, ChatGPT is not without limitations. One potential issue with using ChatGPT is the risk of bias. As with any machine learning model, ChatGPT is only as unbiased as the data it was trained on. If the training data contains biased language or information, ChatGPT may perpetuate that bias in its responses. It is important for developers and users of ChatGPT to be aware of this potential issue and take steps to mitigate it.

Another limitation of ChatGPT is its lack of understanding of tone and context. While ChatGPT is able to generate responses that are appropriate in the context of a conversation, it may not always fully understand the tone or intent behind a message. This can lead to misunderstandings or inappropriate responses in certain situations.

Despite these limitations, ChatGPT is still a valuable tool for assisting with online conversations and generating responses. It is important for users to be aware of its capabilities and limitations, and to use it appropriately in order to get the most out of this technology. In the next section, we will conclude our discussion of ChatGPT and look at its potential future developments.

Conclusion

In conclusion, ChatGPT is a chatbot powered by the advanced language model GPT-3. It is designed to assist in online conversations by generating responses based on user input. ChatGPT has a wide range of potential applications, including customer service, answering frequently asked questions, and providing a virtual conversational companion.

Despite its capabilities, ChatGPT is not without limitations, including the risk of bias and a lack of understanding of tone and context. It is important for users to be aware of these limitations and to use ChatGPT appropriately.

Looking to the future, there is potential for further developments and improvements in ChatGPT technology. As GPT-3 and other language models continue to advance, ChatGPT may become even more sophisticated and capable of handling a wider range of tasks and conversations. It is an exciting time for chatbot technology, and ChatGPT is at the forefront of this evolution.

---

What is ChatGPT? was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

If you’re looking to get ahead of the competition, you need to ensure that you have a competent team with the right mix of professional skills. Many businesses now use ATS systems to find the best candidates more efficiently and easily.

An Applicant Tracking System (ATS) can help you save time and money while finding the best candidates for your business. This guide will discuss how ATS works, the benefits of using an applicant tracking system and how it can help your business grow!

What is an applicant tracking system, and how can it benefit your business?

An applicant tracking system (ATS) is a software that helps businesses manage recruitment, from advertising job openings to tracking applicants and storing employee records. By automating these tasks, an ATS can save businesses time and money.

There are many different types of applicant tracking systems available, and the features they offer vary depending on the specific needs of the business.

There are several benefits of using an applicant tracking system for businesses. First, it can save time by automating tasks that would otherwise need to be completed manually. This can free up time for businesses to focus on other aspects of the recruitment process. Second, an ATS can help businesses keep track of applicants and their qualifications. This information can be useful when making hiring decisions. Third, an ATS can help businesses comply with laws and regulations related to employment. For example, many applicant tracking systems include tools for managing equal employment opportunity (EEO) data.

Overall, an applicant tracking system can be a valuable tool for businesses looking to streamline the recruitment process and make more informed hiring decisions.

How can an applicant tracking system help you find the best candidates more quickly and easily?

An applicant tracking system (ATS) can help you find the right talent easily by automating the process of tracking, screening, and ranking job applicants.

Recruiting can be a time-consuming and expensive process, but an applicant tracking system (ATS) can help streamline the process and save you money while finding the best candidates for your business. Here’s how:

• An ATS can help you post your job openings to multiple job boards with just a few clicks, saving you time and effort.
• An ATS can automatically screen candidates based on criteria you set, so you only spend time reviewing the most qualified candidates.
• It can help you keep track of all your applicants in one place, so you never have to search through piles of paper or emails to find a candidate’s resume or contact information.
• An applicant tracking system can automate repetitive tasks like sending rejection letters, so you can focus your time on more important tasks like interviewing candidates.

So if you’re not using an ATS in your recruiting process, now is the time to start!

What are some of the other benefits of using an applicant tracking system?

An ATS can help to ensure compliance with Equal Employment Opportunity (EEO) laws and regulations. By keeping detailed records of each candidate who applies for a position and those who are interviewed and ultimately hired, an organisation can more easily identify potential patterns of discriminatory behaviour.

Another benefit of using an ATS is that it can help to reduce the risk of hiring errors. For example, suppose a company is looking to fill a position that requires specific skills or qualifications. In that case, an ATS can screen candidates and identify those who meet the desired criteria. This can help ensure that only qualified candidates are interviewed and considered for the position, which can help reduce the likelihood of making a bad hire.

By keeping track of when positions are opened and closed, as well as which candidates are applying for which positions, organizations can better understand where their talent pipeline stands at any given time.

How can you start using an applicant tracking system in your business?

When it comes to an applicant tracking system, there are a few different ways to integrate it into your business process. One way is to use a specifically designed system for tracking applicants. This can be done by consulting a good ATS system provider and getting their product demo to understand if it is easy and cost-effective.

Once you have decided on the type of system you want to use; the next step is to set up the system. This can be done by purchasing an applicant tracking software program or setting up an online account with an applicant tracking service. Once you have the system set up, you will need to add your applicants to the system so they can be tracked. You will also need to set up deadlines for each applicant so that you know when they need to be interviewed and when their applications are due.

An applicant tracking system (ATS) is a software application that helps businesses identify, track, and manage job candidates through the hiring process. ATS systems are used by businesses of all sizes to streamline the recruiting process and improve candidate quality. You just need to understand how ATS works.

If you’re interested in using an ATS for your business, there are a few things you need to do to get started. You’ll need to choose the right ATS as per your needs. There are many types of ATS systems on the market, so choosing one that will work well for your business is important. Finally, you’ll need to train your staff to use the system to maximize its effectiveness.

---

Things You Need To Know About Applicant Tracking System was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

Introduction

The next few decades will be a time of massive change for the global education system. Artificial intelligence is already changing our lives, and it’s also going to revolutionise how we learn. In this article, we’ll look at how AI can improve the way we learn and teach—especially in 2023.

The language and tone-of-voice of our conversational content has a profound impact on the relationship between a modern organization and its customers. This is as true today as it was in the past. Rhetoricians and performers like Plato, Cicero and many others have built their reputations on their ability to captivate an audience. However, where content guidelines for traditional (digital) media could be managed by a relatively simple style guide, conversational AI is much less predictable since it involves real-time user responses and unexpected turns. This 2-way dynamic requires a more holistic approach to content delivery.

As a conversation designer I can still learn a thing or two from this rich discourse of rhetoric, and in this article I’ll investigate what our ancestors thought were the main characteristics of great rhetorical performances.

First point of order is to understand that even though rhetoricians and other performers were speaking to, and not with an audience, they were constantly aware of the effects of their performance on their audience. The delivery of a performance was of the highest importance in the Greek and Roman times, because this meant an orator could influence his audience. The most famous example of the orator’s influence are Cicero’s Catalinarian orations. In a series of speeches delivered on multiple nights he successfully convinced the Roman Senate of Cataline’s plot, saving the Roman Republic from his treachery and systemic collapse.

Through the ages, the rhetorical discourse expanded into the delivery of vocal and instrumental music, stage acting and musicals, and today the advent of conversational AI allows for a new level of emotional connection between users (the audience) and a computer (the performer). Unfortunately the quality of chat and voice experiences isn’t always up to par. As conversation designers it is our job to prevent users from getting frustrated and to reach high satisfaction scores.

Quite a lot of research has been done on the emotional connection between humans and computers. We know that users interact differently with chatbots, voice assistants and other human-like virtual experiences than with a website or app. The technology of Natural Language Programming enables us to create conversational agents with smart, human-like responses that bring about feelings of familiarity and empathy among its users. This phenomenon is called pareidolia. As a result, organizations are starting to recognize that these experiences are ideally suited to convey both a logical answer and build emotional connections between their virtual assistant and their customers.

As conversation designers, we often have to make assumptions as to what a user might say to our assistant. In an ideal world the conversational copy is well-researched, and both happy and non-essential paths are written and implemented before the assistant goes live. But in reality, it doesn’t quite work like this due constraints in time and resources. The modern, lean organization requires us to push our assistant live as early as possible. Secondly, we know that users sometimes prefer to be prompted with buttons rather than an empty line to write on. For these reasons, conversation designers often end up designing an interaction between two people rather than only thinking up our virtual assistant’s lines, and improving it later based on a review of the conversational logs. This approach helps us predict and strengthen the effects of pareidolia with our users, resulting in better experiences.

The above considerations teach us that a holistic approach of content and delivery is essential for a great conversational experience, and lucky for us, the classical rhetoricians were experts in the field. Each period and author has a unique view on what drives a good performance, but in general they agree that 4 principal elements make a good rhetoric performance: Audience and affect, delivery, structure, and ornamentation and repetition. I’ll summarize these below now, and hope to explore each of these in more detail in future blog posts.

Audience and Affect
This element emphasizes awareness of the audience. The performer needs to understand their demographic, language, status, and their number. The goal of this rhetorical element is delectare — to charm and amuse. The language, time and place, style and décor should add to the audiences’ delight.

Affect can be summed up in the French saying “C’est le ton qui fait la musique”, meaning that our tone of voice has a profound impact on our delivery. The tessitura is the relative position of the voice within its natural range, and we can embellish it with intentional intervals and changes in the speed of our delivery. In modern days we see designers of text-to-speech content for voice assistants use different base voices and SSML to create perfect affect.

Allegory and symbolism create strong images in our audience’s minds. In his fourth Catilinarian oration Cicero paints a heroic picture of himself, when he says “…the Gods have determined that he should snatch the senators and the people from miserable slaughter, their wives and children and the vestal virgins from most bitter distress, the temples and shrines of the gods, and that most lovely country of all of them, from impious flames, all Italy from war and devastation.” (reference link, from p.57).

We don’t know if the Roman Senators actually believed this to be true, but the allegory of pain and devastation definitely made it easier for his audience to grasp what the loss of the values and structure of the Republic could mean to them. We don’t need to be overly dramatic — allegories of a more benign nature already have a strong effect on the engagement level of our chatbot users.

Delivery
The delivery, or pronunciatio describes the linguistics and style of written and spoken language. Most rhetoricians considered this the most important skill in a performer’s toolkit. It involves (among others) the performer’s variety, dynamics, humor, physical appearance and articulation. In Conversational AI (CAI) we know that variety in answers adds to the human-likeness of the agent and higher user satisfaction rates. We want a multi-turn conversation with our agent to hit the right tone at every stage. This is where dynamics come in. Humor helps to lighten the mood and when jokes are crafted to fit the brand and bot persona it builds great engagement.

Physical appearance can be translated to the chat UI: is it a widget, is it right in the middle as a part your webpage, or is it an in-app experience? Is it styled in the brand colors?

For our voice assistant it’s probably even more important to understand how looks impact our delivery. Does the assistant live in a home pod with screen, or without? Is it cool and effective or warm and cute? Can it move? Does it have an avatar in 2D, or maybe even in 3D for the metaverse?

Structure
Historically, a good artistic structure consists of three main elements: exordium, narratio, and conclusio. In we typically refer to these elements as the introduction, the statement of facts, and the closing statement. The narratio (statement of facts) is all about instructing the audience of important truths.

The exordium and conclusio are known to make or break a performance. The messages and emotional charge need to be performed with great care so to reach the desired effect. For me the description of these three parts is how conversation designers should approach bots in the FAQ phase. In these conversations we should engage users right from the start, then convey the logical answer to their question, and finally use the closing ‘statement’ to give extra meaning to the conversation and stir brand engagement.

Quintilian (p153) adds three more elements to the mix which are very interesting from a conversation design perspective. First, he mentions divisio, which describes story points to come to the audience. In chat and voice, it’s very important to set expectations before engaging in a long conversation or process with the user. For example, we could start a process by saying “I can help you replace your credit card. I’ll need to ask you 4 questions, and this will take about one minute. Shall we start?” By laying out the next steps in the interaction we foster our user’s understanding and patience.

Confirmatio is another element that Quintilian mentions. This part delivers evidence that helps confirm the audience’s propositions. Providing proof can be very important to maintain trust between an AI and a human. An example could be when a ‘stop smoking’ bot claims that non-smokers live longer. A link to a WHO or NHS statement could help the user’s trust and determination.

Excited, that’s what I’m feeling after going through all these. A virtual assistant is so much more than a tool for automation. The European Renaissance saw a revival of the Greek and Roman classics, recognizing their wisdom. I have seen enough evidence to believe that classical rhetoric will help us design better user experiences and stronger brands. Is it too bold to say that we’re at the starting point of the ‘renaissance of the classics in Conversational AI’? In any case, I will continue to explore how to combine this discourse with the modern skills of conversation design and UX design. Maybe one day, my digital assistants will be as convincing as Cicero in his Catalinarian orations.

Note: Many of the insights in this article derive from ‘The Weapons of Rhetoric’ — Judy Tarling (2004). It’s recommended reading for anyone interested in this topic.

---

How the Western rhetorical tradition strengthens UX of today’s chatbots and voice assistants was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

What is In-Car Speech Recognition?: Definition, use cases and future — General Knowledge

Speech recognition is one of the most exciting technologies in the market today. It has been around for over a decade now. The technology has seen a lot of growth with the likes of Siri and Google Assistant.

Speech recognition is making its way into cars, and it’s one of their most under-utilized features. Given the amount of driving that the average person does, it makes sense to use this technology to make driving safer and more comfortable.

The automotive industry is evolving, and the daily commute is changing. A car is capable of acting as a smart device, and, like other smart devices, speech recognition technology is becoming a commonplace in-car feature. This technology makes it easier and safer for drivers to use their vehicles while remaining focused on the road.

In this blog, we look at the different aspects of the in-car speech recognition market and what the recent trends have been.

What is In-Car Speech Recognition?

In-car speech recognition is a technology that allows drivers to interact with their vehicles using voice commands. This can be used for various purposes, such as making phone calls, sending text messages, playing music, and navigating.

In-car speech recognition systems use training data and a combination of artificial intelligence, machine learning, and hardware to process the dataset and interpret the driver’s voice commands.

The software understands the driver’s intent and converts it into instructions that the car’s systems can execute. The hardware translates the driver’s speech into text.

With the availability of multilingual speech data, automatic speech recognition systems are gaining popularity as they offer a way for drivers speaking different languages to interact with their cars.

There are some concerns about the safety of these systems, as they can be distracting and may not be able to interpret the driver’s commands. The technology is still far from perfect for other languages than English, but is gaining momentum with the availability of data solutions increasing every year.

How Can In-Car Speech Recognition Help?

In-car speech recognition can help reduce a driver’s time looking at a screen or inputting data. It provides a hands-free way to interact with a car’s infotainment system. In-car speech recognition can help to make the driving experience safer and more convenient.

Few of the ways that in-car speech recognition can help drivers include:

• Making hands-free phone calls
• Sending and receiving text messages
• Getting directions and navigation
• Searching the internet
• Controlling in-car entertainment systems
• Opening and closing the car doors

Top Players of Voice Recognition Technology in Automobiles

There are many voice recognition technology providers in the automotive industry. But, there are a few service providers that have become more prominent in the speech recognition industry. Here are the top three players in the voice technology space:

The Future of In-Car Speech Recognition

The future of in-car speech recognition looks very promising. With the advancements in artificial intelligence and natural language processing, the accuracy and usability of in-car speech recognition systems will improve.

According to Automotive World, by 2028, 90% of new vehicles sold globally will have in-car voice control embedded. As the technology develops, in-car speech recognition will likely become even more sophisticated.

In the future, we may see cars that can understand and respond to more complex commands and provide a more natural experience with conversational AI. This could make driving even more efficient and enjoyable.

According to Capgemini, in the future, more people are expected to use voice assistants in their cars. In fact, 73% of people said they would use assistants, compared to the 49% who use them today.

As cars become connected, it’s also likely that in-car speech recognition will become more integrated with other systems in the vehicle. For example, we may see cars that can adjust settings based on the driver’s preferences or that can provide information about the current traffic situation.

Conclusion

New voice technology advancements are here to stay. They will only continue to grow in popularity, making now the perfect time to get ahead of the curve and start creating innovative voice experiences for drivers.

As car manufacturers integrate speech recognition into their cars, this opens up a new world of possibilities for the technology and its users.

Originally published at https://knowledge.world.edu on November 17, 2022.

---

What is In-Car Speech Recognition?: Definition, use cases and future — General Knowledge was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.

Language influences how we see the world. How we see the world influences our language.

Inventing new and alien languages is a hobby that’s been alive for ages, and now, with sci-fi and fantasy shows and movies, it’s a career for some people. More than that, it requires the ability to think outside the box and to challenge sets of rules that we take for granted.

Today, at the very end of 2022, Can an AI invent a new language?

A few months ago I interviewed David Peterson, who creates languages for a living, including for Game of Thrones, Defiance, Dominion, The 100, Penny Dreadful, The Shannara Chronicles, and more. Our conversation went pretty deep (posted at the bottom of the article), so I believed I knew some of what it took to create a new language when I asked ChatGPT to create an alien language.

Here’s what happened.

Request #1: Invent an alien language

Answer: Sure! Here is an example of an alien language I’ve created:

Grammar and syntax:
The alien language uses a subject-verb-object structure for its sentences, similar to English. However, the language does not have tenses, and instead relies on context to convey when an action is occurring. The language also uses a system of prefixes and suffixes to modify the meaning of words.

Vocabulary:
Some example words in the alien language are:

“Zhiv” (verb) — to live

“Zhivik” (noun) — a living being

“Zhivish” (adjective) — alive

“Zhivat” (prefix) — life-related

“Zhivon” (suffix) — relating to the act of living

Sample sentence:

“The Zhivik eats the food.” (Translated into English)

In this sentence, “Zhivik” is the subject, “eats” is the verb, and “food” is the object. The sentence conveys that the living being (the Zhivik) is performing the action of eating something (the food).

Thoughts: That is a pretty good job coming up with an a different concept to base a language on. Well done on that level! Where had ChatGPT been able to pull that from?

But what I really wanted was for ChatGPT to give me more than one word, but a complete paragraph in an alien language.

I tried and failed to get it to do the “to be or not to be” monologue in the invented alien language (Hamlet had been translated into Klingon a few decades ago). But aside from changing two words, the monologue was the same as the original.

I tried again.

Request: Invent another alien language

Answer: Sure! Here is another example of an alien language I’ve created:

Grammar and syntax:
The alien language uses a verb-subject-object structure for its sentences, similar to Japanese. The language also has a complex system of honorifics, which are used to convey respect or politeness when addressing someone. The language does not have tenses, and instead relies on context to convey when an action is occurring.

Vocabulary:
Some example words in the alien language are:

“Gor” (verb) — to do

“Gorik” (noun) — a doer

“Gorish” (adjective) — doing

“Gorat” (prefix) — related to doing

“Goron” (suffix) — relating to the act of doing

Sample sentence:

“Gorik eats food.” (Translated into English)

In this sentence, “Gorik” is the subject, “eats” is the verb, and “food” is the object. The sentence conveys that the doer (the Gorik) is performing the action of eating something (the food).

I hope this helps! Let me know if you have any other questions.

Thoughts: All right. Now the AI’s pattern on ‘inventing langagues’ is clear. I again attempted to get it to write complete sentences in an alien language. Again I failed.

Suddenly I thought of the following request. And it gave me a surprising response.

Request: Write a limerick in an alien language

Answer: Sure! Here is a limerick written in the alien language I created earlier:

Gorish zhivik, with zhivat gor
Ate zhivon food, then zhivat snored
But when zhivik woke
Zhivik’s belly did choke
On all the zhivon food, zhivat deplored

Translated into English:

Doing living being, with life-related do
Ate living-related food, then life-related snored
But when living being woke
Living being’s belly did choke
On all the living-related food, life-related deplored

I hope this helps! Let me know if you have any other questions.

Conclusion: Language goes deeply into how a society thinks, but also into how we humans think. Although at present the AI’s automatic response doesn’t compare to an expert human’s, it is clearly within eventual reach of at least being able to copy their work.

But it does deeper than that.

Having the ability to change basic concepts of language indicates the potential and eventual ability of AI to have concepts different from ours. Not just a little bit different, but wildly different. Not just different in the way that expert mathematicians create concepts that we can conceive of but are hard to understand. But different at a level that’s alien — different in a way that we can’t comprehend, while a brain built unlike ours can.

To achieve true diversity of thought, leading to new and amazing concepts, we may in the future need a brain that isn’t like ours at all.

The future’s here, and the future’s coming. I’m glad I‘m here for that ride.

And for those of you who want to listen to the interview with David Peterson about language and inventing alien languages, check it out here:

---

I Asked ChatGPT to Invent an Alien Language – This Is What Happened was originally published in Chatbots Life on Medium, where people are continuing the conversation by highlighting and responding to this story.