An alien civilization is testing out its magical matter transporter, which can transport random sections of the universe into their lab. One day while testing, they press the button and a dog from Earth appears in the transporter in front of them. The aliens are shocked and overjoyed. This is by far the most interesting thing to ever come out of their transporter. However, due to the random nature of the technology, they cannot keep transporting bits of the Earth to them to learn more. All they have is a 6-year-old golden retriever named Charlie.

The aliens may choose to study the mechanics of our good doggo Charlie to learn more about his providence. They might marvel over the incredibly intricate machinery that makes up each cell, study closely the code of DNA and the clockwork machines of proteins, watch the conscious behavior and response to stimuli, and be impressed by the ingenuity and negative entropy of the object.

The head scientist gathers all of their colleagues around, and presents them with the question: is this dog a result of natural evolution, or was intelligence involved in its creation? How would they know, devoid of context? Would they be able to be confident in any final guess?

The somewhat blurry ground truth here, of course, is that dogs are a naturally occurring form of life that was influenced by intelligence. Would our aliens be able to infer this through some kind of inherent property of our doggo Charlie?

Test #1: Agent Intelligence

An alien scientist suggests that the answer is obvious - Charlie, while lovely, is not very intelligent - even for a dog. Charlie displays no self-awareness, language skills, or complex problem-solving abilities. While it's plausible that Charlie is merely a master of subterfuge and is obfuscating his true intelligence from the aliens, an assessment of the processing capabilities of Charlie's brain seems to show otherwise. Because of this cognitive ability, the scientist proclaims, that such an object is undoubtedly arrived at through natural selection.

But, says another scientist: the low intelligence of single agents within a civilization may not be indicative of the total or even average intelligence of that civilization. As we've previously discussed, intelligent hivemind systems have good reasons why they might produce agents of relatively low complexity. As Charlie possesses a fair few mechanisms for low-bandwidth communication, such a theory cannot be discounted.

Conclusion: A good start, and certainly the easiest option is to just ask the being and trust their answer. Intensive analysis of Charlie's woofs and barks, however, has failed to produce a usable language model to phrase such complex questions.

Test #2: Entropic Analysis

One of the aliens suggests analyzing the level of redundancy and noise within Charlie. For instance by analyzing the DNA to see how much stored meaningful information that did things, and how much had no distinguishable purpose. They reason that a highly intelligent species will only ever produce very efficient and compact objects with very little redundancy or high entropy. Therefore, they claim, the presence of noise within the object is evidence that it is natural.

But, says another scientist: it is impossible to accurately judge the redundancy of Charlie with only limited knowledge of the information. For instance, sections of DNA that the aliens may naively judge to be useless - i.e. they produce no meaningful behavior within a cell when read by normal processes - may be well-encrypted semantic information that the aliens just don't have the decryption keys for. It would not be possible to discern such a scenario. Well-encrypted information is not distinguishable from random noise, so we can't say that "things that look like noise to us" within a system are necessarily indicative of high entropy.

Conclusion: An object may certainly be shown to be low-entropy, with a large part of its contents containing both informational structure and utility. But we cannot make strong claims about an object being high entropy, because we might just not have the right context to be able to access it. Considering also that any form of autonomous object is likely already very low entropy, depending on these small variations seems doomed to fail.

Test #3: Specific and Irreducible Complexity

The aliens wonder about the ability of natural systems to produce certain levels of complexity. The mechanisms that make up Charlie's biology are highly complex and contain many complex biochemical interactions. The systems seem too complex to be from a natural evolution, as they rely on too many interdependent parts that would have to have evolved all in parallel. Surely, this would require someone, somewhere who had an understanding of these processes? What if they were to find highly complex ideas like prime numbers or other mathematical series within Charlie's biology, would that be evidence for design?

But the scientists conclude that this argument is too close to that put forward by some religions of their world, where an intelligent creator made life on their planet to a design. Even if there are highly complex concepts embedded within Charlie's system, this argument depends on a "god of the gaps". It assumes a complexity ceiling on concepts that can come out of emergent systems. If they accept the natural evolution of their form of life and its development of complex chemistry, then they must accept the possibility of other emergent systems that could produce very complex, unintelligent structures.

Conclusion: Complex design may signify intelligence, but it is difficult to define a reasonable boundary above which the complexity of an object becomes too much for a non-intelligent evolutionary process to produce. If it is an insufficient argument to convince the aliens of intelligence creating life on their planet, it cannot be sufficient in other areas.

Conclusion

The purpose of this thought experiment is to be on the other side, where we discover some artifacts of an alien civilization. Not necessarily a creature like our Charlie, of course, but perhaps an object of great complexity. It will be very difficult for us to make any strong claims about whether such an artifact was designed by some intelligence or created through natural evolutionary processes if we have limited context. Even very complex features of such an artifact must be treated cautiously as evidence of intelligence.

The other side of the coin, of course, is the inability to strongly discount intelligence even in very natural-seeming artifacts. Just as Arthur C. Clark once said that sufficiently advanced technology is indistinguishable from magic, we may find that sufficiently advanced technology is indistinguishable from biology.

What do you think? What test would you propose to the scientist? What undeniable traces would intelligence leave upon an artifact? Was this whole thing just an excuse for me to post dog gifs?