Introduce these people and how they contributed to the development of artificial intelligence? Create a chart to show the comparison between your work and ChatGPT’s side by side.
(~50 words each)
Below I have created this infographic in Genially.
The white text blocks are the introductions that I have written myself, whereas the black text blocks are written by ChatGPT3.5.
Click on the photos for links to Wikipedia references of each scholar.
The prompt:
Provide a brief introduction of the following people, and provide an account of what they define as intelligence. Shorten each entry to less than 50 words. Alan Turing, John McCarthy, Herbert Simon, Marvin Minsky. Timnit Gebru
How do “machine (programming) languages” differ from human (natural) ones? (~100 words)
I think one main difference between programming and natural languages is the flexibility in which language can be used, and the goals in which the language is used for. With programming languages, it requires more rigid and structured syntax such that the machine can register and follow through with its given tasks. With human languages, there is more space to flaunt and subvert conventions of syntax, semantics, and pragmatics to be creative, communicate or express a broad range of emotions.
Machine languages, used in programming, are precise and unambiguous with strict syntax, designed explicitly for computer communication. They lack redundancy and prioritize efficiency. Human languages, on the other hand, are flexible, expressive, and context-dependent. They allow for redundancy, creativity, and adaptation to diverse communication needs, serving purposes beyond information exchange. Learning machine languages is a deliberate process for programmers, ensuring computers understand instructions. In contrast, human languages are acquired naturally, evolving over time for cultural expression, emotional communication, and conveying complex abstract ideas with adaptability and flexibility in mind. -- ChatGPT3.5
How does “machine (artificial) intelligence” differ from the human version? (~100 words)
I believe one main difference between artificial and human intelligence is the way that human intelligence can account for a larger range of contexts and nuance, by taking bodily sensations, emotions, and “intuition” into account. A big topic of discussion during my undergrad cognitive systems courses is also whether or not artificial intelligence has “consciousness” or understand certain “qualia” of experiences, given the often seemingly “behavioristic” measures that we can observe artificial intelligence by.
"Machine (artificial) intelligence" differs from human intelligence in essence and approach. While AI excels in data processing, pattern recognition, and repetitive tasks, it lacks human-like understanding, consciousness, and emotional intelligence. AI operates within defined parameters, lacking the intuitive, contextually nuanced reasoning of human cognition. Human intelligence incorporates emotional, social, and moral dimensions, enabling creativity, empathy, and complex decision-making. AI lacks the innate adaptability, self-awareness, and holistic understanding that characterize human intelligence, which integrates sensory experiences, emotions, and a deep contextual grasp for more nuanced problem-solving and learning. -- ChatGPT3.5
How does “machine learning” differ from human learning?
Human learning in general feels more holistic and can take into consideration past experiences that are unique to each individual, which is a continuation of the previous question on the discussion of “qualia”. This integration of personal narrative can change over time as well, and create novel and serendipitous connections to propel forwards in continuation of learning. I am not sure whether or not machine learning can thread experiences over time to inform its own learning.
"Machine learning" differs from human learning in methodology and depth. While machines excel at processing vast datasets, identifying patterns, and optimizing based on predefined criteria, human learning involves intuition, emotional intelligence, and a deeper contextual understanding. Machines rely on algorithms and statistical models, learning from explicit data inputs, often lacking common-sense reasoning or the ability to grasp abstract concepts effortlessly. Human learning is holistic, integrating sensory, emotional, and social cues, facilitating nuanced decision-making, and adapting to diverse situations. Unlike machines, humans possess innate creativity, curiosity, and the capacity for lifelong learning beyond structured training sets.-- ChatGPT3.5
How do your answers to these questions differ from what a machine could generate? (~200 words)
I found that ChatGPT3.5 responses tend to be more general and overarching in a way that provides an overview of the question in a glance, which I appreciate when it comes to wanting to understand concepts and ideas on a surface level. Sometimes when I read ChatGPT responses, it feels very emotionally “flat” and has a strange “diplomatic” tone that makes the content of the response feel rather sterile and impersonal, but more objective and factual at the same time.
In comparison, my human answer tends to be more focused on a few aspects that delve a little deeper, with anecdotes that refer to my personal experiences. My choice of words to use would feel more or less with usage of words that seem more “alive” and less “uncanny” (I hope!) which makes it is something that ChatGPT3.5 is unable to provide.
References
OpenAI. (2024). ChatGPT (Jan 24 version) [Large language model]. https://chat.openai.com
Timnit Gebru, (2024, January 16). In Wikipedia. https://en.wikipedia.org/wiki/Timnit_Gebru
Harris, J. (2023, May 22). “There was all sorts of toxic behaviour”: Timnit Gebru on her sacking by Google, AI’s dangers and big tech’s biases The Guardian. https://www.theguardian.com/lifeandstyle/2023/may/22/there-was-all-sorts-of-toxic-behaviour-timnit-gebru-on-her-sacking-by-google-ais-dangers-and-big-techs-biases
John McCarthy, (2023, December 05). In Wikipedia.https://en.wikipedia.org/wiki/John_McCarthy_(computer_scientist)
McCarthy, J.(2007, November 12)What is Artificial Intelligence? https://www-formal.stanford.edu/jmc/whatisai/whatisai.html
Marvin Minsky, (2024, January 04). In Wikipedia. https://en.wikipedia.org/wiki/Marvin_Minsky
Herbert Simon, (2024, January 24). In Wikipedia. https://en.wikipedia.org/wiki/Herbert_A._Simon
Alan Turing, (2024, January 24). In Wikipedia https://en.wikipedia.org/wiki/Alan_Turing
Hi Sophy,
I’ve decided to produce this weeks’ feedback through an AI avatar for a gauging way for you to experience the feedback pertaining to artificial intelligence. Please let me know if you have any questions. To start, you do an excellent job at providing an excellent overview of the key players in the research, design, and execution of what we know artificial intelligence to be today. As you compare this with the chat, GPT version, you can see your version is more elaborate and detailed. I appreciate you connecting to the Wikipedia documents. Just make sure that you’re always referencing your work and cross checking with multiple sources.
As you investigate how machine languages differ from human languages, you speak to the flexibility in the language. Machine languages are definitely more structured from an academic standpoint and as we know, there is no tone within machine generated language. The affordance of human language on the other hand is to make critical connections based on past experiences and backgrounds which machine language cannot compute. Although humans can still find a scholarly feel to their work, machine language will never match the authentic quality of human language. This is the section where you discuss “flaunting subversions”. Coming up with a personalized example to connect with this point would be a great next step.
Your discussion about human versus machine intelligence is excellent. I love that you draw on an experience from your undergraduate degree. Make sure that if you were using full quotations that you’re quoting something from scholarly work, if it’s not from an academic source, just use a single quote. With languages, human intelligence is impacted by internal and external circumstances, experiences, and meaning making. This has a profound impact on the way humans think versus the way they code machines. As machines are programmed by humans, they succumb to the biases that humans portray which is a large flaw to the computer generated language. Humans have these innate biases, but because human intelligence occupies a vast amount of individuals and perspectives, human biases can be overcome, transformed, dismantled and resolved through a single interaction in fact, whereas this is much harder to achieve in machines.
I very much appreciate your argument about human learning occupying personal narratives that can change and expand over time, which are not necessarily true for machine languages. Humans are influenced on cultural, social, local, and individual levels, and as such learn in multiple ways based on their multiple intelligences and affinities. The same does not hold true with you Machine learning. I would love for you to do a deep dive into this idea, obviously within the confines of the word count. What example might you be able to give to encapsulate this idea?
As you explain, the responses provided by ChatGPT as they adhere to machine learning, language, and intelligence are from more of a generalist perspective and lack personalized experiences and emotion that are central to humans. How might you create a conclusion that encapsulates the overall outcome of your learning within this assignment?
Sophy, you’ve done a great job with comparing these questions using your response versus the ChatGPT. Well done. Please let me know if you have any questions.