Logical reasoning PrepTest 147 · Section 1 · Question 24

Incorrect. The argument isn't that true AI is impossible but rather that doing so requires more than capturing the information in the human genome. Therefore, this answer that it's "impossible" to capture these processes is out of scope.

Correct. Argument or Facts:
Argument

Valid or Flawed:
Flawed

Question Type:
Strengthen with Necessary Premise

Stimulus Summary:
Computer Scientists Ð To make AI, all we need to do is make a program with the info from the human genome
Biologist Ð The brain's operation is governed by proteins encoded by the genome, so you can't just capture the genome to make AI

Answer Anticipation:
Heady stuff. And yet another question in this section with a lot of technical/scientific concepts!

The Biologist here concludes that the Computer Scientists are mistaken, so it can be helpful to rephrase that conclusion to feature the content of what the Biologist believes to make it easier to see any gaps. Their mistaken belief? That an AI can be made by capturing all the information in the human genome. So the Biologist, in finding that to be mistaken, believes that it's not enough to capture all the information in the genome to make an AI.

Why not? Because the brain works through protein interactions. So while these proteins have structures encoded in the genome, the Biologist must believe they also contain other information not in the genome—their interactions must be governed in some other way.

The correct answer, then, should states that the interactions of these proteins aren't determined by the genome, and thus simply mapping the genome isn't enough to replicate the function of the brain.

Answer Explanation:
This answer reflects the assumption made by the Biologist. If the interaction of these proteins is encoded in the human genome, then it's possible that AI could be generated just with that information. The argument falls apart if this isn't true, so it's the correct answer.

Key Takeaway:
Whenever the main point of an argument refers back to an opposing point, make sure that you rephrase that conclusion to reflect the content of the belief. For instance, it was a lot easier here to see the gap in the logic when thinking about the conclusion as—"Making a program that contains the info in the human genome isn't enough to make an artificial intelligence"—rather than thinking of it as—"The computer scientists are mistaken."

Incorrect. The debate here is over whether modeling an AI on the human brain via the human genome would work, not whether it's the only method that would work, so this answer is out of the scope of the debate.

Incorrect. The Biologist's argument doesn't rely on the computer scientists's method being too difficult, but rather to miss some relevant information, so this answer isn't necessary.

Incorrect. The Biologist's argument is that the computer scientists are missing the point, not that they're going for the easy way out. Even if the program that contains the info from the genome is harder to write, the Biologist's argument about what is really required to get the functioning of the brain can still hold.