Note: I can't comment on how this is generally approached in other subjects, but you're referencing Einstein, so as a physicist, I feel semi-qualified to weigh in on how we do it in physics. Note that this isn't necessarily general advice, but it likely applies elsewhere.
"Einstein et al.'s ..." works OK, although in the majority of occassions I can think of, I'd rewrite to get rid of the possessive; "Einstein's et al. ..." is definitely not something I would use or have ever seen or heard of.
Details and examples
There are a few possible cases/options that apply here: While this list isn't necessarily comprehensive, you might want to...
- use a formal, 'standard' referencing convention, in which case, every reference perhaps looks like something between square brackets e.g.
[Mit21] or possibly an author (with et al.) and year e.g.
(Rincewind et al., 2021) - no-one (that I know of/have seen) uses possessives of these, so it's not part of the 'standard convention', so the only way forward would be to reword the sentence.
The gedankenexperiment ['thought experiment'] in  demonstrated ...
- use an informal standard referencing convention. This is as per 1. but a bit more personal. I'd generally think of this in the first person, but in certain contexts, I see no reason it wouldn't work in the third person as well.
In , we considered the use of ...
mix the notation given in the question with standard citation. I'd consider this to be the third person version of 2. although could also be made more formal.
Cubitt et al.  proved ...
For more details, see the proof in/by Cubitt et al. 
stick with 'et al.' but remove possessiveness. Could be useful if the paper and authors have already been mentioned. Would still recommend a proper citation.
Squarepants et al. experimentally demonstrated that living underwater ...
use the possessive form of 'et al.'. I have only ever seen this as "et al.'s" but even this is relatively rare and I would generally recommend one of the other options above.
Hong et al's experiment in the 1980s was...
Generally, the closer to the top of the list, the more formal the expression but overall, the preferred expression depends a lot on the context and desired emphasis. If you're writing a presentation for a small group of people you already know, being more informal might actually be a good thing. If you're doing a presentation about the history of certain experiments, you're likely going to want a better way to refer to them than a number, so informally referring to them by author et al.'s names (i.e. just like that) would work quite well in this situation. However, if writing a paper intended for submission to a journal, it's likely that more formal expressions are more appropriate.