For instance the problem with high energy physics is not that people don't have ideas but rather we have limited experimental and observational results that would let us tell one theory from another.
Similarly I don't think more intelligence would help in solving the Collatz conjecture.
Maybe too much?
Both extents are consistent with Sutskever's ideology that Lossy Compression [Reduction?] Is all U Need (probably superseding AIaUN)