MB6 — Selection and Basin Stability

Percolation-style cooperation evidence is assumed to warrant basin stability (MB6a), and a stable socio-technical basin is assumed to support correction integrity rather than select against it (MB6b).

What decision changes?

Before trusting that an aligned equilibrium will persist, ask whether the basin is stable because it is good, or merely because it is stable — a bad basin can be just as sticky.

Most alignment agendas hold a single system fixed and ask about its weights. This bridge does the opposite: it makes selection by deployment leverage (which systems institutions copy, fund, and deploy) and basin stability load-bearing, on the claim that outcomes depend on which systems institutions select, not on any one system’s internals alone.

MB6 is really two linked bridges. MB6a says percolation-style cooperation evidence is enough to call a basin stable. MB6b says a stable basin actually supports correction integrity, rather than merely being stable. That second step is the sharper bet: value lock-in is a direct counterexample to MB6b, because a stable basin can be a stably bad one.

The book’s own formal spine flags a further weakness: if this bridge’s evidence and MB8’s legitimacy-theater check both route through the same self-report or cooperation signal, treating them as two independent lines of defense is a mistake — a single steerable chokepoint can block both at once. No experiment has yet shown that the two instruments are independent in practice.

What would count as evidence?

Evidence would include tracking whether deployment-leverage selection pressure (which systems institutions copy, fund, and deploy) preserves or erodes correction-channel integrity over time, independent of any single system's weights.