Who were these people, and how did/does each contribute to the development of artificial intelligence?
How do “machine (programming) languages” differ from human (natural) ones?
The main purpose of language is to communicate- the way in which machines and humans perform this task differ completely, though both are logical with their own devised syntax and structure. Arguably, human language is more complex with underlying emotions and nuances involving non verbal cues and inferential reasoning for understanding (Jones, 2020). Machines are programmed by humans with defined rules that do not deviate from its algorithm. As such, machines make meaning that are distinct to the words, such that each command has one meaning. Jones (2020) suggests that machine language is performative such that the code executes its function to perfection whereas human language can take on a variety of meanings given the context.
How does “machine (artificial) intelligence” differ from the human version?
“Artificial” being the key difference between machine and human intelligence. One is created by and mimics humans with the input of information. Human intelligence requires cognitive processes that are both unconscious and subconscious, whereas machines require direct instruction for processes (Dreyfus, as cited in Crawford, 2021). Chollet (2019) theorizes that AI can be measured by skill acquisition efficiency in comparison to a more in depth measure for humans, the Intelligence Quotient, that considers brain volume, speed of neural transmission and working memory (Stangor & Walinga, 2014). The unpredictable nature of the human brain involves the ability to think whereas AI is dependent on datasets.
How does “machine learning” differ from human learning?
Humans learn from experience and we learn from one another. Alternatively, machines are trained by processing a large set of data and noticing patterns within the dataset to make predictions (Heilweil, 2020). Inherently, both humans and machines face biases, the differences being in how they are learned. Algorithmic biases exist based on the biases of the programmers who design them as well as the bias in the dataset. Machine learning lacks foresight as the information that informs the actions are based on historical and past data.
And for your LAST challenge, a version of the Turing Test: how do YOUR answers to these questions differ from what a machine could generate?
In theory, a machine would be able to generate answers that are similar to these, however there are a few differences between ‘us’, whether it be a positive or negative attribute. The most obvious being the time spent on these responses. A machine would be able to generate answers fairly quickly with its vast availability of data online; whereas the time I spent was drawn out across several hours in the week. There would also be discrepancies with the accuracy of the responses, notably in the style of writing, the formatting requiring APA and the references to other sources. Notably, I chose a multimodal format to display the biographies, whereas a machine would likely choose a conventional written format, which is also efficient. Lastly, my personal background, experience and knowledge contributes to the answers, where the machines would rely on data across the web.
References:
Chollet, F. (2019, November 5). On the measure of intelligence.
Crawford, K. (2021). Atlas of AI. Yale University Press. (Introduction: pp. 1-21)
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.
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.
Stangor & Walinga, 2014. Introduction to Psychology – 1st Canadian Edition, part of the B.C. Open Textbook Project