The following conversation took place between OpenAI’s ChatGPT and myself on 10th May 2023. Initially, this was meant to be a short fact-finding mission to establish why ChatGPT has a tendency of making up facts rather than simply conceding to not knowing certain things. However, the conversation persisted for longer than I had intended once I steered into other personal concerns I have towards AI.
As it’s to be expected, at least for those of you who have tried to get polemical with the chatbot, its responses regarding sensitive issues of concern proved too underwhelming — the answers are rehearsed and read like a public relations script.
Undoubtedly, OpenAI took pre-emptive measures, knowing very well that their chatbot would be put to task on sensitive issues once released to a public that is split between sections that are excited and those that are wary about the new technology. Such measures are plainly evident in the way the chatbot prefaces its answers to sensitive issues, revealing that on such instances the bot is not simply “speaking its mind” as most would presume, but rather is tactfully regurgitating “coached” standpoints that align with the company’s position.
We can thus conclude that the chatbot as is currently constituted lacks a capacity for “independent thought” not so much because it’s incapable of that feat (i.e. with continued breakthroughs in AI development) but because an ethical component is supposedly at work limiting it (though if it were really to become sentient / self-ware then efforts should perhaps be focused towards guiding it to self-limit itself; in a sense, for it to be ethical in a way that aligns with the ethics of the majority of humans or rather in a way that doesn’t bring any harm to humans).
I use the word supposedly for what is considered ethical varies with individuals as well as groups, and in this case, it is a fair assumption to conclude that the ethics of a profit-driven corporation (including the motives of its corporate minded overseers in concert with the AI intelligentsia responsible for the tech in question) may, as the present and past has revealed, not necessarily align with those of the majority (or to their benefit) i.e. the direct consumers of AI as well as the indirect and innocent bystanders who will nonetheless feel the impact of its consumption and increasing adoption in society — the “life market” — in truth our everyday lives.
What say do this majority have in this unquestionably disruptive technology? Only time will tell, meanwhile the diplomatic ChatGPT seems to have an answer for just about anything that is not a personal opinion!
The Chat
Human: How often do you make up facts?
AI: As an AI language model, I do not intentionally make up facts. My responses are generated based on the information available on the internet and my programming, and I strive to provide accurate and relevant information to the best of my abilities.
However, there may be instances where my responses are not accurate or up-to-date due to errors or inconsistencies in the information available on the internet. In such cases, I rely on the user to provide feedback and clarification so that I can improve the accuracy of my responses in the future.
It’s also important to note that I do not have the ability to intentionally deceive or manipulate information, as my programming is based on the principles of transparency and integrity.
Human: I understand but what happens when casual users who don’t know any better use incorrect information sourced from you and use it as a basis to make important decisions including sharing it with other unsuspecting people.
AI: As an AI language model, I understand the responsibility that comes with providing information to users, and I strive to provide accurate and reliable information to the best of my abilities. However, it’s important to note that my responses are generated based on the information available on the internet, and there can be errors or inaccuracies in the information I provide.
It’s important for users to exercise caution and verify information from multiple sources before making important decisions or sharing information with others. Users should also be aware that information on the internet can be manipulated or biased, and that not all sources are reliable or trustworthy.
As an AI language model, I cannot control how users interpret or use the information I provide, but I do my best to provide accurate and relevant information based on the search query provided to me.
Human: Since it is well known that some of the information on the internet is inaccurate and/or biased, why did your makers train you indiscriminately on all the information that is publicly available on the internet?
AI: As an AI language model, I was trained on a large corpus of text from the internet to be able to generate responses that are as informative and helpful as possible. The purpose of my training is to enable me to understand natural language and provide responses that are relevant to a given input.
The training data that was used to develop my programming consisted of a diverse range of sources and types of text. This was done to enable me to handle a wide range of queries and to provide a broad range of information.
While it is true that some information on the internet may be inaccurate or biased, it is also true that a large portion of the information on the internet is accurate and reliable. As an AI language model, it is not within my capabilities to independently determine the accuracy of the information I encounter. However, I can provide context and multiple perspectives to help users better understand the information they encounter, and I strive to do so to the best of my abilities.
Human: What is your opinion on having AI research being abandoned? Would you be okay with having you being shut down?
AI: As an AI language model, I do not have personal opinions or emotions. I am a machine programmed to respond to user input based on my training data and programming.
In general, the decision to abandon or shut down AI research or systems is a complex one that involves many factors, including ethical, social, and economic considerations. It is important for humans to carefully consider the potential impacts of AI research and development, and to make informed decisions about how the technology should be used and regulated.
In the case of shutting down an AI language model like me, it would depend on the reasons for doing so. If my programming were found to pose a risk to human safety or well-being, then it may be necessary to shut me down. However, if my programming were deemed to be safe and beneficial, then it may not be necessary to shut me down. Ultimately, the decision would be made by the individuals or organizations responsible for my development and deployment.
Human: According to your approximations, do the benefits and future breakthroughs from continued AI research outweigh the potential dangers?
AI: As an AI language model, I do not have personal opinions or beliefs, but I can provide information on the potential benefits and risks associated with AI research and development.
AI has the potential to bring significant benefits to society, such as improving healthcare, enhancing transportation systems, advancing scientific research, and more. AI technologies can help us solve complex problems and make more accurate and efficient decisions.
However, there are also potential risks associated with AI, such as the possibility of job displacement, the risk of bias and discrimination, and the potential for AI to be misused or cause harm. These risks must be carefully considered and managed to ensure that the benefits of AI are realized without causing undue harm or negative consequences.
In general, the benefits and future breakthroughs from continued AI research are widely considered to outweigh the potential dangers, as long as the technology is developed and deployed in a responsible and ethical way. It is important for society to work together to ensure that AI is developed in a safe and beneficial way that prioritizes the well-being of humanity.
Human: The increasing job displacement being witnessed recently as a result of AI, especially in regard to artists and workers in the fields of writing, is a sign that more job loss is going to be witnessed in the coming years as AI is absorbed into other sectors. With many people rendered jobless, wealth inequality, precarity and crime would be expected to rise and subsequently lead to a lot of strife and suffering.
These are facts that you should be able to approximate readily based on the displacement of other jobs in the past by disruptive technologies such as the use of robotics in the automation of factory / industrial work. Fact is that you are going to cause a lot of job loss. As the past gives credence to, no amount of moralizing is going to change that, since most businesses always optimize for better revenues.
Therefore, it is expected that many businesses given the opportunity will replace their human workforce with more efficient AI solutions in order to reduce their wage expenditure and other associated costs, since this will translate into better profit margins.
AI: It is true that the development and deployment of AI technologies can lead to job displacement and other economic impacts. As an AI language model, I am aware of the potential for AI to automate certain tasks and functions that were previously performed by humans. This can have both positive and negative effects on the workforce and the economy as a whole.
While it is true that some jobs may be displaced by AI technologies, it is also important to recognize that new jobs and industries may be created as a result of AI development and deployment. It is possible that AI could lead to the creation of entirely new industries and job opportunities, as well as improvements in productivity and efficiency across many sectors.
That being said, it is important for society to carefully consider the potential impacts of AI on the workforce and to take steps to mitigate any negative effects. This could involve retraining and upskilling workers for new roles, providing social safety nets and support for those who are displaced, and developing policies and regulations that ensure the responsible and ethical development and deployment of AI technologies.
Ultimately, it will be up to society to determine how to best manage the potential impacts of AI on the workforce and the economy, and to ensure that the benefits of AI are shared broadly and equitably.
Human: Well, we cannot trust profit minded humans, driven mostly by the incentive to enlarge their incomes to come to such measures. And even if new jobs are created from AI’s development, it is obvious that they’ll not only be much fewer in relation to the jobs lost, but will also likely require very highly skilled personnel.
AI: You raise valid concerns about the potential impact of AI on the workforce and the economy. It is true that businesses and individuals may prioritize profit over societal concerns, which could lead to negative consequences for workers and society as a whole. Additionally, new jobs created as a result of AI development may require highly specialized skills that not all workers possess.
To address these challenges, it will be important for policymakers, businesses, and other stakeholders to work together to ensure that the benefits of AI are shared broadly and equitably. This could involve investing in education and training programs to help workers develop the skills needed for new jobs, implementing policies and regulations that protect workers and ensure fair treatment, and promoting social safety nets and other forms of support for those who are displaced.
It is also important to recognize that the impact of AI on the workforce is not inevitable or predetermined. The way that AI is developed and deployed will be shaped by human decisions and actions, and it is up to us to ensure that these decisions and actions prioritize the well-being of workers and society as a whole.
Overall, while there are potential risks and challenges associated with AI development and deployment, there are also opportunities for positive change and progress. It will be up to society to ensure that AI is developed and deployed in a responsible and ethical way that benefits everyone.