Robert Shapiro's 1986 book "Origins" had a rather snarky takedown of that kind of OoL research.

The research does something like this. You run an experiment and show some chemicals are present in low concentration. You say "ah ha, we've shown one step". Those chemicals are now taken off the shelf in high concentration and used in the next experiment, which makes slightly more complex chemicals in low concentration. Repeat this multiple times until done.

Lampooning this, the fictional OoL researcher in the book also demonstrates that monkeys can type out "To be or not to be, that is the question" on a typewriter. He arranges the typewriter with an automatic right margin and lets them type away until he gets a line starting with "T". First letter demonstrated! Now he rereun the experiment, but now with paper the "T" in the first column and margins set to the second column. Repeat this with subsequent letters until the entire sentence is produced. Of course, the monkeys didn't type the sentence, the researcher did.

Getting around the exponential complexity barrier is the whole goddamn point, it's not something to just wave away with a "we don't know" comment.

IMO some fundamentally new idea is needed for OoL.