“Meaning” sounds as creationist to me as “purpose”, but it’s also clear that David is talking about this from a non religious standpoint. I think he’s saying that by studying natural systems and identifying their functions (which correspond in my mind, to their “purpose”), we can copy and adapt them to perform other functions, according to our wishes, or purposes. After all, all human functional design has a goal or purpose.
I see this as just another facet of evolution, competing (perhaps) with other means of evolving. We are ultimately part of the whole that evolves (at least in one way of explanation).
I’m struck by the level of resistance to David’s approach that has appeared on this thread, especially when it is not just the basis of his proposal for network updates, but as we all know, it is fundamental to every aspect of SAFE Network operation! Did you not get the t-shirt? 
I agree it raises questions that we should explore. It is certainly breaking new ground, and will be hard to achieve. But the apparent fear that the approach will fail assumes that a human top down design, is somehow more robust, because we attempt to address a set of goals in a more methodical, piecemeal, reductionist, manner. When we already know that this approach is never robust. That such designs always go through long periods of testing, tweaking, fixing, and often major revisions, and that this process never stops. Just look at the security problems of the current internet, which has become a dog’s breakfast of top down designed cludges that constantly fail as each new security hole and other faults are discovered and exploited.
How is it, that in the face of so much evidence that top down systems designs, useful and ubiquitous as they are, constantly fail and need to be fixed, that we jump to defend it with the implied assumption that a bottom up rule based approach, modelled on the most amazing designs we know (ecosystems) will necessarily be worse. Especially when it is already fundamental to the success of SAFE Network!
Yes, it’s new, untried, risky, and may fail. But that’s evolution! And evolution rocks IMO.
If one wants to talk about meaning and purpose in the metaphysical sense, I think evolution is worth much study. I’m not advocating going there, just suggesting we don’t need to be afraid of, or to see such considerations as dangerous, or inconsistent with a pragmatic approach to delivering SAFE Network. They are valid questions too, as are all questions from a scientific perspective, and to me, shutting the door on them is unnecessary, and just as “dangerous” as any kind of fundamentalist thinking.
Even if it were caught early before total network collapse, somebody’s valuable data is still at risk even with redundancy. I don’t know about you, but I hold all of it in high regard. None of it is expendable. Handing over the entire network and its data to an A.I with the intelligence equivalent to that of a preschooler is not advisable. Only when matured and proven should the A.I even begin to be considered for seeing the light of SAFE 
Though still not recommended. This takes years of rigorous testing. Trying to implement it any sooner is careless and irresponsible. Truly deserving of a nuclear bomb up the a@@.
