- Lowlife builds are "fine". I think some top-end nerfs need to happen, but they don't need to be drastic. It's the lack of diversity in other archetypes and the continued building of synergy into the archetype that make the problem. It's important to recognize that Lowlife builds are actually generally two very different build archetypes. Lowlife Flawless and Lowlife Iceborne.
Lowlife Flawless, which uses discipline and predator in conjunction with lowlife, is mostly fine. It has a VERY demanding ask of the player's mastery of the game to use correctly. Its main problem is not itself a consequence of its own power, but of many, many factors extrinsic to it that allow it to stack power on power. And even if you nerf the component perks, the remaining archetypes are very close behind in terms of power and hungry to take its spot, without any of other power components that drove it specifically over the edge also being taken out. And since the goal is to take out the power problem holistically (I hope), nerfing Lowlife Flawless components outright along with those other driving factors may be an overcorrection. I'd be very careful when addressing these components directly because they are not specifically the major cause of the power problem, and they become much healthier when those other causes are controlled better due to things like staggerlocking and multiplicative damage sources having a disproportionate effect on outcomes when paired with this raw power. There is also just the simple problem of relativity. Lowlife Flawless is only so outstandingly powerful relative to the power floor. If we had methods of having a higher power floor, then the content floor could be lower. If the content floor were lower, then Lowlife Flawless wouldn't seem so stupendously outperforming.
Lowlife Iceborne on the other hand, namely the iceborne portion of it, has its problem rooted in "you can make infinite mistakes and always recover". Iceborne has no limiting factor that punishes you for making repeated mistakes or limits the amount you can recover from your mistakes. The 50% hp (which is actually meaningfully higher ehp due to the DR%) "restriction" is not a meaningful restriction. My personal preference for iceborne is turning it into a red/chip health mechanic of sorts; if you take damage, half of it or so is turned to red/chip health. You can lifesteal back the red/chip health, but if you get hit again, you lose any remaining red/chip health you didn't heal back. I understand that it would be a lot of work, particularly on the UX/UI side of things, for a mechanic that only happens with Iceborne, but since Iceborne is so widely used, I think it makes sense to invest in that direction slightly.
This is the important bit: I think most folks problem with Lowlife builds isn't with Lowlife builds itself. Their problem is that there's no alternatives which compete, and that even if you don't build lowlife out of principle, your build will play exactly like lowlife. That's the problem at the kernel of the critique, I believe - not that lowlife's too strong, just that there's no meaningful alternative. People want to play builds with a greater amount of identity in them, to feel like even if they built an optimized build, it wouldn't play exactly the same and so they're getting an intrinsic reward out of that playstyle, and that they have a greater amount of options. As I said in my critique a few days ago, even small changes in playstyle identity have a massive impact on the psyche of players. Boreus Momentum Blades CB becoming a thing with a simple mechanics change has had a much larger impact on the identity and play of CB users than many of the deliberate and work-intensive development choices. It speaks volumes that players have such a disproportionate visceral impact to something as small as "which buttons do I spam" in the scheme of things compared to large design decisions. And I don't mean to say "stop making large design decisions" by this - rather, I'd like to think you can learn more about how to affect your community with your design decisions with this information so you can make smarter use of the bandwidth you have. Players are strange and find rewards in things developers might overlook the importance of.
I'd like to add a small tangent here and just provide what might be a controversial take... I don't think Attack Speed should be a mutable stat. Or at least not one so easily mutable. I think some mechanics, be it from the behemoth, or a weapon's toolkit, etc. etc. can make sense to enhance attack speed. But at its core, I think attack speed is so inherently core to combat design (giving you, the developer, the tools to say "what kind of attacks should fit in what kinds of openings for what kinds of situations, etc. etc.,") that giving players easy ways to subvert that combat design is unhealthy outside of very carefully weighed and measured scenarios and toolkits. Its impact on damage output is also significant, having a multiplicative damage output in addition to its possible effects on "break points" - the idea that you can fit X number of combos in an opening, but with attackspeed you might be able to fit X+1 number of combos into an opening, an important thing in a game where much of your damage is often backloaded into combo finishers on every weapon - making it hard to ignore its role in the power creep issue over the beta where we had very few attack speed options but many such options in the current iteration of the game. Base attack speed is also core to many weapons' identities. Being able to speed up an axe to chainblade-levels of chop-chopping is funny and novel, but it also undermines the core fantasy of the axe. This, on top of many input-related bugs with attack speed leads me to the feeling that attack speed should NOT be a stat players can so freely alter. If weapons don't feel good at the base attack speed, change the base attack speed, don't lock making the weapon feel good behind RNG cell acquisition, and absolutely do everything you can to keep a tighter lock on combat. The reason MH's combat is so beloved is because of its sincerely deliberate attention to how weapons engage in fights, and attack speed is possibly the most visible way that it manages to build weapon identities at a core level, as it defines how a weapon class even begins to engage with a fight.
In conclusion, I honestly am at a loss as to what exactly you could do in specifically the short term, lol. There's so many systemic changes and neglected overgrowths to be trimmed that simple and immediate fixes are unlikely to do a whole lot to fix the crucial things that need to change. I think your biggest challenge is going to be identifying the things which are fine to keep, which need to go, and not overcorrecting or undercorrecting based on those perceptions. And I think smart decisions will have a more significant impact on the community more than large decisions. It's a very difficult task you've got for yourself, but one which I think is entirely feasible to solve in a way that addresses the concerns of the community, newbie and veteran alike.