Yes, I guess there are two main scenarios here. 1) Simple translations that google translate or similar can do sufficiently well. This exists, is useful in man

Yes, I guess there are two main scenarios here. 1) Simple translations that google translate or similar can do sufficiently well. This exists, is useful in many scenarios, such as reading signage or posters and other stuff, but I never use it IRL because I don't want to walk around looking through my phone all the time. And for traffic signs, I obviously can't go throuh all of that while driving. AR can solve this when the tech is small enough. It's not so much a case of that I'm too embarrassed to pick up my phone, it's more that the expected value of any single translation is very low, but taken all together I'm missing a lot of information, some of which could be useful. 2) Complex translations of documents and such. I have tried google translate for this but the text is typically both too complex and also can't fit on the screenspace so it never manages to get the whole paragraph to translate. Some of this could theoretically be solved with a super high res AR device with a great translation AI, but obviously then a similar tech on my phone would also do the trick.