This is the era in which artificial intelligence (AI) has become more prevalent and is increasingly replacing mid-level skills across both the developed and developing worlds (Majumdar et al., forthcoming). This is also the era in which data privacy is becoming more regulated and the ability of the individual to own his or her own data a reality through technologies such as blockchain (Verbert, Sharples and Klobucar, 2016). In this new normal, one has to consider how learning, delivered through multiple platforms and modes, can be credible, authentic and transferable. Traditional forms of quality assurance, many of which are closely linked to the pervasive development of national and regional qualifications frameworks across most parts of the globe, have played a key role for the last 30 or more years (Allais, 2010; Allais, 2017; Allais, Raffe and Young, 2009; Allais et al, 2009; Branka, 2016; International Labour Office, 2017; McGrath, 1997; Raffe, 2009), drawing on many decades of models and systems that preceded qualifications frameworks. The emerging view is that quality assurance is even more necessary in this new era of multiplicity and diversity, but also that it needs to evolve. Quality assurance in the new world will be more private sector-driven, more open and more digital. This will include industry and vendors as the supply and demand value chains are completed in more direct, and to some extent also unpredictable, ways than have been possible to date. The object of assurance will be less a qualification and more a credential. New quality dimensions that distinguish between presage, process and product will become commonplace (COL, 2016), so too quality labels that signal benchmarking through an emphasis on self-assessment, and lighter touch approaches (European Association of Distance Teaching Universities, 2016).