A larger union is likely to bring its own set of office politics and costs.
Just remaining within the absolute letter of the law is complex and panties for failure can be heavy - I know I have dealt with this aspect in the UK.
- Employees are regularly screwed on misrepresented equity compensation
- Employees are expected to work uncompensated on-call shifts which vary from "doesn't affect me" to "impossible to function on" such as being paged once per hour for 1 week with the expectation to do it again in a weeks time.
- Employees are subject to highly opaque performance review processes that take into account employee performance, as well as what work they happened to be assigned - Some assignments may be incompatible with the companies expectation for a given role e.g. Software Engineers assigned to put together excel reports
- Employees may be asked to perform ethically or legally dubious activities where they may be held criminally liable ( see dieselgate, and to a lesser extent the 737 max ).
- Employees may be asked to work 80+ hours to hit goals which were underfunded or poorly planned (see CDPR and every other software project which misses a deadline)
This is a small list of reasons an industry employee group such as a guild/union/certification board/association could be a good idea. However very few of these items relate to traditional union activities. The thought of pulling in specific hires with specific union skills seems like either forcing a square peg through a round hole or just paying to support someone who will fill their time with activities related to "unions" to little employee benefit.
I have seen a lot of cases relating to "PRP" and Performance reviews.
Realistically AWU has to partner with another union initially standing up a brand new stand alone union with all the costs and difficulties is not practicable.