Tailwind CSS Navbar component
The Imposter Among Us
A Description of a game and extension of the Turing Test
Authored by Mohamad Khawam - April 06, 2025
The Imposter Among Us
Tags:
AIGameTuring Test

Alan Turing's is a inspriation of mine and I have always been fascinated by the ability to tell wether something is human or not. The Turing Test is an experiment proposed by Alan Turing in 1950 to determine whether or not a machine can exhibit intelligent behavior which is indistinguishable from that of a human. The test involves a human interrogator who interacts with both a human and a machine through a computer interface such as a console (A. M. Turing, 1950). The interrogator must then determine which of the two entities is the machine and which one is the human based solely on their responses to questions. If the interrogator is unable to reliably distinguish between the two, the machine is said to have passed the Turing Test.

In recent Jounrals, LLM such as GPT-4.5 are said to be able to pass the Turing Test (Jones, Bergen 2025). However, I believe that the Turing Test demonstrates a single persons ability to distinguish a response as being from a human. This is not the only test of intelligence, as it relies on the subjective judgement of the interrogator. The test is conducted using a single question and answer format, which does not accurately reflect the complexities of human communication and intelligence. When we communicate with each other, we do not simply ask questions and wait for answers from other another. We also predict the anwser we will get given a certain response. We also take into account the context of the conversation, such as the other person's tone, body lanaguage, and facial expressions. The Turing Test measures a single aspect of intelligence, which is the ability to respond to questions in a human like way. But, is far from the only aspect of intelligence required for a machine to be considered intelligent. Let alone be able to exsist in a human society.

Humans are social creatures, and our intelligence is deeply intertwined with the social interactions we have with each other. We learn from the interactions we have with one other. Our abilities to not just understand each other, but grow with each other is what makes us human. The Turing Test does not take into account these nuances of human communication which are essential for a machine to overcome before being accepted into society.

There's a few questions that need to be asked before a machine can be considered intelligent. For example, can a machine understand the context of a conversation? Can a machine understand the emotions of the people its interaction with? Can a machine learn from its interactions with humans and adapt its behavior accordingly? These questions provide the basis of machines ability to be accepted into society. A machine that cannot understand the context of a conversation will struggle to communicate effectively with humans. A machine that cannot understand the emotions of people will struggle to build relationships with humans its interacting with. And most importantly, a machine that cannot learn from its interactions with human will struggle to adapt and grow with humans as it interacts with them.

The imposter among us

The imposter amoung us is a simple game where humans interact with machines in a social setting. A human and a machine are placed in a chat room with a few other humans. The interagator just like in the Turing Test, is tasked with determining which of the players is the machine. As the conversation progresses, the interrogator is meant to be part of the conversation and interact with the other players as they see fit. This creates a more dynamic and complex environment for the machine to navigate. The interrogator is not just asking questions, but also responding to the other players in the chat room. This allows the interrogator to get soical feedback from the other players in the chat room. The interrogator is not just a passive observer, but an active participant in the flow of the conversation. The goal of the machine is to blend in and join the flow of the conversation. Is the machine able to continue the conversation as if its a another human? The interrogator must determine which players are machines and which are humans based on their responses in the chat room. If the interrogator is unable to determine which player is a machine and which is a human, then the machine has successfuly be able to imposter itself as a human. It has demonstrated the skills required not only to learn from the conversations its having, but able to continue with the flow of conversations between humans.

Why conversation between mutliple humans?

The Turing Test is conducted in a small experiment with a single human and a machine. The interrogator is tasked with determining which of the two is a machine. This limited experiment does not take into account the complexities of human communication and what it takes to continue a conversation with other humans. The goal of the Turing Test was to tell whether or not a machine has the skills to take tasks ordinarily done by humans and replace them. But, does not take into account that many tasks done by humans are done based on social context. For example, is a machine able to divert from the topic of conversation or change the subject? This are skills used by humans to navigate conversations. When done with a single person it might be easy to change topics, but doing it with the context of many players can be challenging. Having a machine be "within" a conversation with other humans allows the interrogator to see all aspects of human communication. And how a machine will act within that coversation.

Conclusion

The Turing Test is a valuable experiment in the field of artificial intelligence, but it is not the only test of intelligence. The proposed game, "The Imposter Amoung Us," provides a more complex communicational environment for machines to have to navigate. The game allows for a social setting where machines must be able to learn and adapt to the flow of conversation. The question becomes can a machine replace your friend?

Sources

  • A. M. Turing. (1950). Computing Machinery and Intelligence. Mind, 59(236), 433-460.
  • Cameron R. Jones & Benjamin K. Bergen (2025). Large Language Models Pass the Turing Test.