I don’t claim #2.
What I do claim is that when we think natural laws are producing new MI, then it will always turn out on further analysis the MI came from somewhere else where it pre-existed. Chance and determinism can never produce a net increase in MI. This is just the law of information non-growth.
So the following is correct:
And per a private communication with @swamidass:
Replication does seem to depend on prior MI. Basically, there is no way to get MI for free, so in general any instance of MI is indicative of ID
Consequently, fine tuning is also an instance of the information argument, but the information argument is more general and applies to more areas. This is why I said all the ID arguments reduce to Dembski’s CSI argument. CSI and the information producing capability of intelligence is the central issue.
It becomes scientific rather than merely philosophical in the case of ID claiming we can detect the creation of new information and distinguish this from information being passed on from another source.
To bring this back to DNA and evolution, some make the claim that evolution can originate the MI in DNA. In fact, that is its whole claim to fame, that natural selection and mutation are sufficient to create all the genetic diversity we see, as if there were some special MI creating ability in the evolutionary process itself. But, this we now know to be false due to the law of information non growth.