The past number of years have seen peak excitement, enthusiasm, and inflated optimism around AI, mostly predicated upon yet to materialise “benefits”, “potentials”, and “promises”. Actual deployed systems however continue to fail to live up to their promises, leading to numerous downstream impacts including erosion of public trust in AI. In this talk, we survey current state-of-the-art AI systems and their downstream impact on society and present recommendations geared towards reliable and equitable AI systems as well as mechanisms for accountability.