Wait until people start training AI on AI output because (a) that's the only source of information left with a large enough corpus size and (b) all other sources are polluted.
There's no way of asserting accuracy when we reach that point because the very nature of information and scientific thinking will be corrupted.
There is evidence to suggest that LLMs have recently passed a threshold where their performance can be improved by training on their own output.
https://arxiv.org/abs/2210.11610
I expect your concerns are likely unwarranted.
How does this do anything but validate their concerns?
They're not saying "if we train LLMs on their own output, they'll get slower or worse", they're saying "if we train LLMs on their own output, it gets even harder to actually derive sourcing for what they produce."
That's what proponents of self-driving cars keep promising also.
You can’t run benchmarks on an AI-clean meat sack.