And any mechanical engineer would tell you this is where a ton of issues begin - when you start joining/combining materials next to each other when they have different Young's modulus/physical properties etc. If you were to weld a 2ft aluminum bar to a 2ft carbon steel bar to form a 4 feet long bar and then flex it, the ways that both materials resist that flexion amplifies stress at the join. Even in a mechanical join (like riveting/any other form of bonding - heck even doing a friction weld or casting) the weak point is the join. It appears like this craft had a titanium "nose" that was connected to a carbon fiber hull. It doesn't even matter if both materials on their own were rated to handle the 6000 PSI pressure down there, the fact they would handle that pressure differently creates the opportunity for massive failure.
Especially with respect to carbon fiber - what's really difficult to know is when potential failure points start propagating - even when you x-ray it (the only way to see internal faults in CF). For something that goes under pressure to that insane degree, it doesn't really matter how many layers you use - it will gradually start failing and the pressure just amplifies that. A Carbon fiber bike doesn't last forever for this reason - it also doesn't undergo 6000PSI. I would hypothesize this is why it could make a handful of trips without technically "failing" - but the stress on key points of the hull/construction was causing rapid degradation with no way of assessing that. The last trip was one time too many.