Who were these people, and how did/does each contribute to the development of artificial intelligence? How did/does each think “intelligence” could be identified?
| My Responses | Chat-GPT | |
| Alan Matheson Turing | A British mathematician, Turing is famous for his participation in codebreaking the German Enigma machine (Biography, 2020). Turing, known for the Turing test, contributed to artificial intelligence because it raises the question, “Can machines think?” (Turing, 1950). The Turing test determined if humans could detect computer answers from human answers. | A pioneering British mathematician and computer scientist, Turing laid the foundation for AI with his Turing Test, a measure of a machine’s ability to exhibit human-like intelligence. His work on the Enigma machine during World War II and the Turing machine concept profoundly influenced computer science. |
| John McCarthy | Coining the term artificial intelligence, McCarthy invented LISP, a programming language for AI (John McCarthy, 2019). McCarthy believes intelligence to be “the ability to achieve goals in the world” (Sutton, 2020, p. 66). For example, this could include people or machines such as a thermostat, as they can achieve goals. | An American computer scientist, McCarthy coined the term “artificial intelligence” in 1955. He founded the Stanford AI Lab and developed the Lisp programming language, a key tool for AI research. McCarthy focused on creating machines capable of reasoning and problem-solving, advancing AI as a distinct field. |
| Herb Simon | Simon developed the general problem solver (GPS), and the logic theorist, believed in demystifying thinking. Simon believes that human intelligence and information processing have three common operations: scanning for patterns, storing patterns in memory, and applying patterns (Frantz, 2003, p. 271). | An American cognitive psychologist and computer scientist, Simon, along with Allen Newell, developed the “Logic Theorist,” the first AI program. Simon emphasized the importance of problem-solving and decision-making processes in AI. He won the Nobel Prize in Economics for his work on bounded rationality. |
| Marvin Minsky | Minsky, a mathematician and computer scientist, built the first neural network learning machine. Minsky, along with McCarthy, cofounded the Artificial Intelligence Project, a research centre at MIT (Dennis, 2019). He believed “the brain is a machine whose functions could be replicated in a computer” (BBC News, 2016). | An American cognitive scientist, Minsky co-founded the MIT AI Lab and made significant contributions to AI, coining the term “perceptron.” His research focused on neural networks and machine perception, guiding the field’s early development. Minsky advocated for understanding human cognition to replicate intelligence in machines. |
| Timnit Gebru | Gebru, a leader in AI ethics, has done work in identifying AI bias, and the harms of AI in terms of discrimination. She has identified four main risks of large language models; environmental and financial costs, massive data, inscrutable models, research opportunity costs, and illusions of meaning (Hao, 2020). | An Ethiopian-American computer scientist and AI ethics researcher, Gebru contributed to addressing bias and ethical concerns in AI. Co-founder of the Black in AI community, she worked at Google on fairness and accountability. Gebru emphasized the need for inclusive, ethically grounded AI, urging responsible development and deployment. |
How do “machine (programming) languages” differ from human (natural) ones?
Programming languages differ from human languages through a variety of means. The first revolves around the study of morphology. For example, in human languages we can often alter the meaning, and context of language using stems, root words, prefixes, and suffixes. In programming, this does not exist. The second major difference revolves around the strict rules programming languages follow. Programming languages were invented with the thought of logic, precision, and being irrefutable, on the contrary, human languages are full of slang, accents, or vernacular that is subject to variations based on context and culture. This does not exist in a programming language.
How does “machine (artificial) intelligence” differ from the human version?
Chollet (2019) identifies two main characteristics of intelligence; one focuses on accomplishing task-specific skills, and the second focuses on the generality and the ability to adapt to various environments. Task-specific skills rely heavily on logic, routines, and developing a large knowledge base (p. 5). Chollet shares the example of AI chess; for humans to become proficient in chess, they need to use general intelligence to acquire a specific skill. However, when a machine is optimized for learning, it is built specifically to handle a narrow task and focus in on a specific task. Whereas generalization, defined as “the ability to handle situations (or tasks) that differ from previously encountered situations” (Chollet, 2019, pp. 9-10), tends to aim to mirror human cognitive abilities.
How does “machine learning” differ from human learning?
One of the main differences between machine learning and human learning is the ability to understand bias. Machine learning relies on being exposed to large data sets of information to judge and predict the information it receives (Heilweil, 2020). Jones (2020), identifies how Facebook used machine learning to place people into “micro-categories” and produces targeted posts to reach niche groups of people. We have seen how this can create gender and racial bias. Digital literacy skills are becoming more invaluable, to differentiate between the biases and determine how algorithms work.
Turing Test: how do YOUR answers to these questions differ from what a machine could generate?
My answers to the above questions would differ from what a machine can produce to some degree. Of course, in my writing you can see citations and my writing style (with likely the odd grammar error). However, regarding the content within the answers, in the chart’s first question, you can see very similar answers between the Chat-GPT and my responses. This could be because the answers to these questions are surface-level questions, able to be found directly by a Google search, or through some reading of some articles. I would be interested to see the remainder of the questions and how the answers are compared. I would think the answers would result in a similar response. This is because the questions do not allow for personal opinion, but rather a statement of facts that an AI program like Chat-GPT could recreate. By adding a personal or cultural context to the questions I am asking, I believe that the answers would become more differentiated and varied.
References
BBC News. (2016, January 26). AI pioneer Marvin Minsky dies aged 88. BBC News. https://www.bbc.com/news/technology-35409119
Biography. (2020, July 22). Alan Turing – education, movie & quotes. Biography. https://www.biography.com/scientists/alan-turing
Chollet, F. (2019, November 5). On the measure of intelligence.
Dennis, M. A. (2019). Marvin Minsky | American scientist. In Encyclopædia Britannica. https://www.britannica.com/biography/Marvin-Lee-Minsky
Frantz, R. (2003). Herbert Simon. Artificial intelligence as a framework for understanding intuition. Journal of Economic Psychology, 24(2), 265–277. https://doi.org/10.1016/s0167-4870(02)00207-6
Hao, K. (2020). We read the paper that forced Timnit Gebru out of Google. here’s what it says. MIT Technology Review. https://www.technologyreview.com/2020/12/04/1013294/google-ai-ethics-research-paper-forced-out-timnit-gebru
Heilweil, R. (2020, February 18). Why algorithms can be racist and sexist. A computer can make a decision faster. That doesn’t make it fair.Links to an external site. Vox.
John McCarthy (computer scientist). (2019, October 15). Wikipedia. https://en.wikipedia.org/wiki/John_McCarthy_(computer_scientist)
Jones, R. H. (2020). The rise of the Pragmatic Web: Implications for rethinking meaning and interaction. In C. Tagg & M. Evans (Eds.), Message and medium: English language practices across old and new media (pp. 17-37). De Gruyter Mouton.
Sutton, R. S. (2020). John McCarthy’s definition of intelligence. Journal of Artificial General Intelligence, 11(2), 1–100. https://doi.org/10.2478/jagi-2020-0003
Turing, A. (1950). Computing machinery and intelligence. Mind, LIX(236), 433–460. https://doi.org/10.1093/mind/lix.236.433
Curtis Bourassa 