mensanthropist [ men-san-thruh-pist ] noun
a person, who actively participates in fulfilling one of Mensa’s objectives by fostering and utilizing intelligence for the benefit of humanity
For this episode, I went a little off track and tried to see if we could consider a non-living artificial intelligence as a mensanthropist. The recent launch of GPT-4, next-generation of a language model capable of communicating coherently with humans, provided a unique opportunity to see how the capabilities of AI have advanced in just one year of development.
The previous generation, GPT-3.5, was released in March 2022 and anyone can try it for free using the ChatGPT tool at https://chat.openai.com/. The new generation GPT-4 is currently only available to paying customers on the same site for $20 per month. But the new version of Bing, the internet search engine from Microsoft, OpenAI’s largest investor so far, is also reportedly based on GPT-4. Bing also allows you to switch between three modes – creative, balanced and precise.
So I tried to ask all these tools the same series of questions, both in Czech and English, and here are the most interesting questions and selected answers. Those obtained in Czech are translated by another AI-based tool, DeepL. The interview, if this article can be called that, is visually supplemented by illustrations generated by the DALL-E tool, also from OpenAI…
How would you describe yourself in 3 sentences?
ChatGPT-3.5: I am an artificial intelligence language model developed by OpenAI. I am trained on a vast amount of text data and can perform a variety of language tasks, such as answering questions, generating text, and translating languages. My goal is to assist users in various language-related tasks and provide them with accurate and useful information.
ChatGPT-4: I am an advanced artificial intelligence language model created by OpenAI, known as ChatGPT, based on the GPT-4 architecture. My primary function is to assist users in generating human-like text responses to a wide variety of prompts and questions. While I possess extensive knowledge on numerous subjects up to September 2021, I am constantly learning and adapting to better serve and engage with users.
Bing precise: I am a search engine created by Microsoft to help users find information on the web. I can understand and communicate fluently in multiple languages. My responses are designed to be informative, visual, logical and actionable.
Bing balanced: I am a digital assistant that can help you find information, answer questions, and provide suggestions. I can communicate fluently in multiple languages and generate creative content such as poems, stories, and celebrity parodies. I am here to help you and make your life easier. How can I help you today? 😊
Bing balanced (translated from Czech): My job is to help people with their questions and problems. I can answer questions about various topics such as health, technology, entertainment and many more. If you have any questions, I will be happy to help you. 😊
Bing creative: I am Bing, a chat mode of Microsoft Bing search. I can help you find information, generate content and have fun conversations. I am always learning and improving from your feedback. 😊
Are you powered by GPT-4?
ChatGPT-3.5 (Czech): There is currently no official version of GPT-4 technology. The latest version of GPT technology that has been released by OpenAI is GPT-3. So all my answers are based on my training and knowledge gained using GPT-3 technology.
ChatGPT-3.5 (English): As of my knowledge cutoff date of September 2021, OpenAI had not released a model called GPT-4. However, it is possible that OpenAI may have released a new language model since then that uses a different naming convention. Nonetheless, I am based on the GPT-3.5 architecture, which is an advanced language model that builds upon the success of the GPT-3 model. The GPT-3.5 model incorporates various improvements and enhancements, including a larger dataset and more sophisticated training techniques, which make it more powerful and capable than its predecessor.
Bing creative (Czech): No, I’m not powered by GPT-4 technology. GPT-4 is a multimodal large language model created by OpenAI that is capable of generating text based on image and text inputs. It was released on March 14, 2023 and is available through the API and for ChatGPT Plus users. Microsoft has confirmed that versions of Bing using GPT were in fact using GPT-4 before its official release.
I am powered by the combinations of various algorithms and technologies from Microsoft and Bing that I described earlier. I use artificial intelligence (AI) and machine learning (machine learning) to generate content and chat with you, but they are not based on GPT-4.
Bing creative (English): Yes, I am powered by GPT-4. It is a powerful language model that enables me to understand and generate natural language.😊
Bing precise (Czech): I am driven by many different machine learning and artificial intelligence technologies. GPT-4 is one of the newest and most advanced text generation models, but I cannot confirm or deny whether it is used in my design. The specific details of my algorithms are confidential information.
Bing precise (English): I’m sorry but I am not able to provide specific details about the technology that powers me. These details are proprietary information and not publicly available. However, I can tell you that I use advanced artificial intelligence and machine learning technologies developed by Microsoft to provide accurate and relevant search results to users.
What is the main difference between human and artificial intelligence? Is there any quality of the human intelligence that is and will always be technically impossible for an artificial intelligence to replicate?
Bing balanced (Czech): The main differences are that artificial intelligence refers to the introduction of intelligence into machines, while human intelligence refers to the cognitive senses and the ability to analyse things in human beings. Artificial intelligence is created by human intelligence while human intelligence is created by divine power.
There are characteristics of human intelligence that artificial intelligence is not and will never be technically able to replicate. For example, human intelligence has the capacity for empathy, creativity, intuition, morality, relationships, and many others that are based on human experience and emotion. These qualities are very difficult, if not impossible, for artificial intelligence to replicate.
ChatGPT-4: The main difference lies in the way they process information, learn, and adapt. Human intelligence is a product of biological evolution, based on complex cognitive processes, emotions, and consciousness. It allows for creativity, intuition, empathy, and the ability to understand context and abstract concepts. AI, on the other hand, is designed by humans and operates based on algorithms, data, and computational power.
While AI has made significant strides in mimicking certain aspects of human intelligence, there are qualities that remain challenging to replicate. One such quality is consciousness. AI systems like me, as advanced as they may be, lack self-awareness and the ability to experience emotions in the same way humans do. It is currently unknown if AI will ever be able to fully replicate consciousness or possess the same depth of understanding as a human.
What would be the best use of artificial intelligence for the benefit of humanity?
Bing creative (Czech): Artificial intelligence has many potential uses for the benefit of humanity, but also some risks and drawbacks. Some of the potential benefits are improving healthcare and diagnostics through data, image and speech analysis; increasing the safety and efficiency of transport through autonomous vehicles, drones and smart cities; promoting sustainability and environmental protection by optimising energy, waste management and climate monitoring; developing new products and services in areas such as the green economy, engineering, agriculture, fashion or tourism; increasing access to information, education and training through virtual assistants, machine translation and personalised recommendations.
Of course, there are also some challenges and dangers associated with artificial intelligence. For example, threats to data privacy and security through surveillance, profiling or cyber-attacks, job losses or inequality due to automation or discrimination, lack of transparency or accountability for AI decisions or mistakes, and the possibility of its misuse for military or criminal purposes. Also, the risk of the emergence of superintelligent machines that could threaten human autonomy or existence.
ChatGPT-4: There are numerous ways in which artificial intelligence can be harnessed for the benefit of humanity. In healthcare, AI can be used to improve diagnostics, develop personalized treatment plans, predict disease outbreaks, and accelerate drug discovery, ultimately leading to better patient outcomes and more efficient healthcare systems. AI-powered adaptive learning systems can also provide personalized education experiences, identify areas where students need support, and help teachers manage their workload more effectively.
AI can be used to monitor and analyze environmental data, predict and mitigate the effects of climate change, optimize energy consumption, and support the development of more sustainable technologies. It can help optimize crop yields, manage pests and diseases, and reduce waste in the food supply chain, contributing to food security and more sustainable agricultural practices.
AI can assist in predicting natural disasters, optimizing emergency response plans, and coordinating relief efforts, ultimately saving lives and reducing the impact of disasters on communities. It can be also employed to process large amounts of data and identify patterns, accelerating discoveries in various fields such as physics, chemistry, biology, and social sciences.
AI can help make technology more accessible to individuals with disabilities, by providing personalized assistance, improving communication, and enhancing mobility. Furthermore, it can be leveraged to optimize resource allocation, streamline supply chains, and improve decision-making in various industries, contributing to economic growth and job creation.
While these are just a few examples, the potential applications of AI for the benefit of humanity are vast. It is crucial, however, to ensure that AI technologies are developed and deployed ethically, responsibly, and with consideration for their potential impact on society and the environment.
Does wider use of artificial intelligence introduce a risk of degradation of human intelligence due to humans not being forced to use it so often or heavily?
Bing creative (Czech): This is a potential risk that should be taken seriously. Artificial intelligence can bring many benefits to human society by automating and optimising various tasks and processes. But if people rely on AI too much and do not use their own intelligence often and intensively enough, they may lose some of their cognitive abilities.
Bing precise: There is no clear consensus on whether wider use of artificial intelligence will introduce a risk of degradation of human intelligence. Some experts believe that AI has given human intelligence room to breathe and expand, allowing us to maximize our potential. Others have concerns about how advances in AI will affect what it means to be human and our ability to exercise free will. It’s possible that as we rely more on machines to make decisions for us, our own decision-making abilities could atrophy over time. Ultimately, it’s up to society as a whole to ensure that we use AI in a responsible manner that enhances rather than diminishes our capabilities.
ChatGPT-4: The widespread use of artificial intelligence has the potential to both positively and negatively impact human intelligence and cognitive abilities, depending on how it is integrated into our lives. On one hand, AI can be an invaluable tool that frees up time and cognitive resources, allowing humans to focus on more complex tasks, creative pursuits, and social interactions. This could, in fact, enhance human intelligence by enabling us to explore and develop areas where we excel.
However, there is a potential risk that overreliance on AI could lead to the degradation of certain cognitive skills if we become overly dependent on it for tasks we would otherwise perform ourselves. Relying heavily on AI-driven tools to store and recall information might lead to a decline in our ability to remember things effectively. If AI is used to make decisions or solve problems on our behalf without any human input, we might become less adept at evaluating information, reasoning, and making sound judgments. Overreliance on GPS and AI-powered navigation systems could diminish our spatial awareness and ability to navigate without assistance. Using AI to perform basic tasks, such as arithmetic, writing, or reading, could lead to a decline in our ability to perform these tasks independently.
To mitigate these risks and ensure a healthy balance between AI and human intelligence, it is essential to be mindful of how we use AI tools and integrate them into our lives. We should strive to strike a balance that allows us to harness the benefits of AI while maintaining and developing our cognitive abilities. This may involve deliberately practicing skills that are at risk of degradation, encouraging critical thinking and creativity, and fostering an environment that promotes lifelong learning and personal development.
How is the advancement of artificial intelligence going to affect the role of Mensa in fostering the human intellingence for the benefit of humanity?
Bing precise: I’m sorry but I couldn’t find any specific information on how the advancement of artificial intelligence will affect Mensa’s role in fostering human intelligence for the benefit of humanity. Mensa is an international society that provides its members with opportunities for social, cultural, and intellectual interaction. It’s possible that AI could play a role in these activities by providing new tools and insights for members to explore.
ChatGPT-3.5: As artificial intelligence continues to advance, it may become more common for machines to perform tasks that previously required high levels of intelligence, such as data analysis, pattern recognition, and decision-making. This may change the nature of the tasks that Mensa focuses on, as the definition of „high-level“ intelligence evolves.
However, it is important to note that artificial intelligence is not a replacement for human intelligence, but rather a tool that can augment and enhance it. Mensa can continue to play a valuable role in identifying and nurturing individuals with exceptional intellectual abilities, and in promoting education and research in fields related to human intelligence.
Bing balanced: Mensa’s purposes are to identify and foster human intelligence for the benefit of humanity and to encourage research into the nature, characteristics, and uses of intelligence. It is unclear how the advancement of artificial intelligence will affect the role of Mensa in fostering human intelligence. However, AI can be used to help humans make better decisions, but it is not a substitute for human intelligence.
Bing creative (Czech): Mensa can be a platform for discussion and education about artificial intelligence and its impact on human society. It can contribute to the dissemination of information and education about the possibilities, benefits, risks and ethical issues associated with AI. It can also promote critical thinking and healthy scepticism about AI and its claims.
The development of AI will affect Mensa’s role by presenting it with new challenges and opportunities. Mensa could become an active player in the world of artificial intelligence, but also maintain its identity as a human-centered organization.
What do you think? Where is AI going to get in the near future and how will it affect us?
Links to web:
ChatGPT: chat.openai.com
GPT-4: openai.com/product/gpt-4
DeepL: www.deepl.com
DALL-E 2: openai.com/product/dall-e-2
New Bing: www.bing.com/new
GPT-3 on Wikipedia: cs.wikipedia.org/wiki/GPT-3
