It's true fill welds are stronger than fusion, I've talked to several welding engineers who have stated the same.
The fusion weld will result, like SF claims, in metal drawing from the parts being joined, metal will sink toward the center of the silencer which without a filler material will result in a slight deficit of material at the point of weld. Another drawback of fusion welding is that penetration is limited to about .035" however that's not a major problem with AAC's silencer as that is the thickness of the tube wall being joined- the resulting thickness of materials joined is probably .025-.03"
Given I believe both products depend on the strength of a front and rear full circumferential weld to hold together, I think those welds are the deciding factor here, and the answer to the question of which is stronger lies in comparison of these characteristics:
Comparing the strength of the fusion circumferential weld by AAC, and the circumferential fill weld by surefire, and firing both silencers to failure to determine whether some other engineering feature AKA baffle thickness/bore diameter and warpage imparted by each of the individual construction processes [IE likelihood of strikes at high barrel temperature]. The ability of welds, baffle design, and mounting system to withstand damage imparted by baffle strikes with the most minimal reduction in overall performance. Different baffle styles may be more and less able to withstand baffle strikes.
This sounds like a problem easily solved with a simple side by side test conducted to a standard like the original endurance test that I've heard the Surefire silencer passed but can't substantiate for lack of a documented source.
That said, I'm pretty sure the Surefire circumferential fill weld is the superior of the two. I think their baffle is more tolerant of strikes, as can be seen in SMG Lee's photos, and I think their 600+ rounds of endurance to failure is going to be very hard to beat if that's the case, which seems to be substantiated by AAC's biased testing and statements in court, in conjunction with Surefires statements in court.
I found this interesting:
M16/M4A1 barrel failure testing
The military testing makes fairly clear that one would need a SAW or HK416 with it's superior hammer forged barrel and gas piston system to even conclude such a test as the M4A1 barrel fails at about 580-600 rounds with no silencer, and attaching a silencer would doubtless reduce this by at least 30% due to longer gas dwell time.
If the silencer is strong enough to destroy the weapon prior to failure, I see no reason for any greater strength.