The Turing test is a test proposed by the British mathematician and computer scientist Alan Turing in 1950. It is designed to evaluate a machine's ability to exhibit intelligent behavior that is indistinguishable from that of a human. The test involves a human judge engaging in a natural language conversation with a machine and a human, without knowing which is which. If the judge cannot reliably differentiate between the machine and the human based on their responses, the machine is said to have passed the Turing test. ChatGPT
If you think about this description of the Turing Test, you might immediately detect the definition's fatal flaws. The flaws are with the human participants. What if the two human participants do not speak the same language or have approximately the same level of intellectual ability? This problem has been addressed academically by discussions involving the question of when a machine becomes intelligent. There is no question that some machines can do things faster and better than any human. Here are some links to articles discussing different aspects of the intelligence question.
- “Can Machines Be Intelligent? | Psychology Today.” n.d. Accessed June 27, 2023. https://www.psychologytoday.com/us/blog/the-intelligence-report/201103/can-machines-be-intelligent.
- “MIT School of Engineering | » When Will AI Be Smart Enough to Outsmart People?” n.d. Mit Engineering (blog). Accessed June 27, 2023. https://engineering.mit.edu/engage/ask-an-engineer/when-will-ai-be-smart-enough-to-outsmart-people/.
- Quora. n.d. “What Exactly Is Machine Intelligence?” Forbes. Accessed June 27, 2023. https://www.forbes.com/sites/quora/2019/11/15/what-exactly-is-machine-intelligence/.
- Vincent, James. 2016. “What Counts as Artificially Intelligent? AI and Deep Learning, Explained.” The Verge. February 29, 2016. https://www.theverge.com/2016/2/29/11133682/deep-learning-ai-explained-machine-learning.
- “What Happens When Machines Become Smarter than People? | Philosophy Break.” n.d. Accessed June 27, 2023. https://philosophybreak.com/articles/what-happens-when-machines-become-smarter-than-people/.
- “Will Machines Become More Intelligent Than Humans?” n.d. Caltech Science Exchange. Accessed June 27, 2023. http://scienceexchange.caltech.edu/topics/artificial-intelligence-research/machines-more-intelligent-than-humans.
What does ChatGPT think about the question?
The question of when a machine becomes intelligent is a topic of ongoing debate and discussion. Intelligence itself is a complex and multifaceted concept, and there are different ways to define and measure it.
In the field of artificial intelligence (AI), machines are considered intelligent when they can perform tasks that typically require human intelligence. This can include capabilities such as learning from data, reasoning, problem-solving, recognizing patterns, and making decisions.
Traditionally, AI has been divided into two main categories: narrow or weak AI, and general or strong AI. Narrow AI refers to systems that are designed to perform specific tasks or functions with a high degree of competence, such as image recognition or natural language processing. General AI, on the other hand, refers to systems that possess the ability to understand, learn, and apply knowledge across a wide range of tasks similar to human intelligence.
Currently, most AI systems in use are considered narrow AI, where they excel at specific tasks but lack the broader cognitive abilities associated with human intelligence. However, as technology advances, there is ongoing research and development aimed at achieving general AI.
The exact point at which a machine can be considered truly intelligent is subjective and may vary depending on the criteria and perspective used. Some argue that intelligence requires consciousness and self-awareness, while others focus on the ability to exhibit human-like cognitive functions regardless of consciousness.
It's important to note that there is no consensus on a definitive threshold for machine intelligence, and the field of AI continues to evolve with ongoing research, advancements, and new discoveries.
It is interesting that ChatGPT did not refer to the Turing Test at all. Can I tell that the text above was written by a computer program? How is using ChatGPT different that looking up articles, like I did above. Is the fact that the computer program in looking at the same data was able to give cogent response make any difference? Tellingly, the computer program did not quote any of the abundant source articles on the subject. By the way, I consider the ChatGPT response to be overly wordy and superficial.
But what about the Turing Test? My answer would be it is no longer relevant. Another note, there are now computer programs that can tell if what is written above was written by ChatGPT or was the original work of a person. See the following:
“ChatGPT Detectors in 2023 - PC Guide.” n.d. Accessed June 27, 2023. https://www.pcguide.com/apps/chat-gpt-detectors/.
As I understand it, AI works by collecting published opinions nn the subject being discussed, This is a very competent summary of what I think about AI,so itseems that most people who are considering this subject, thnk as I do - and I do consider myself quite intelligent. Interestingly, AI has not paid any attention to cospiracy theorists, who do abound on the internet.
ReplyDelete