It is now about 30 years ago that photometric and spectroscopic surveys of asymptotic giant branch ( AGB ) stars in the Magellanic Clouds ( MCs ) uncovered the first examples of truly massive ( > 3 - 4 M _ { \odot } ) O-rich AGB stars experiencing hot bottom burning ( HBB ) .
Massive ( Li-rich ) HBB AGB stars were later identified in our own Galaxy and they pertain to the Galactic population of obscured OH/IR stars .
High-resolution optical spectroscopic surveys have revealed the massive Galactic AGB stars to be strongly enriched in Rb compared to other nearby s-process elements like Zr , confirming that Ne ^ { 22 } is the dominant neutron source in these stars .
Similar surveys of OH/IR stars in the MCs disclosed their Rb-rich low-metallicity counterparts , showing that these stars are usually brighter ( because of HBB flux excess ) than the standard adopted luminosity limit for AGB stars ( M _ { bol } \sim - 7.1 ) and that they might have stellar masses of at least \sim 6 - 7 M _ { \odot } .
The chemical composition and photometric variability are efficient separating the massive AGB stars from massive red supergiants ( RSG ) but the main difficulty is to distinguish between massive AGB and super-AGB stars because the present theoretical nucleosynthesis models predict both stars to be chemically identical .
Here I review the available multiwavelength ( from the optical to the far-IR ) observations on massive AGB and super-AGB stars as well as the current caveats and limitations in our undestanding of these stars .
Finally , I underline the expected observations on massive AGB and super-AGB stars from on-going massive surveys like Gaia and SDSS-IV/APOGEE-2 and future facilities such as the James Webb Space Telescope .