Artificial intelligence (AI) is becoming integral to organizational strategy-practices and, hence, to society. As organizations progressively employ “computational agents that act intelligently” (Poole & Mackworth, 2017: 3) to make vital decisions regarding their operations, society becomes shaped by the judgements of intelligent machines. From The Washington Post’s use of AI to cover the 2020 US elections to Norrtälje Municipality’s use of AI to pre-assess child abuse cases.
The augmented integration of AI is particularly observable in organizations’ strategizing of communication. That is, in strategy-practices that incorporate artificially intelligent communication technology (AICT) in the planning, execution and monitoring of an organization’s purposeful use of communication to fulfil its mission (Edwards, 2012; Hallahan et al., 2007). The airline KLM, for example, use AICT to perform automated and personalized 24/7 customer service; the public relations company Omnicom Media Group, utilise AICT to monitor employees’ communication work; and, rather infamously, the now-defunct consulting firm Cambridge Analytica, employed AICT to design political campaigns.
The society-level challenges and risks of increased AICT adoption have been highlighted in a number of recent publications (e.g., Araujo et al., 2020; Flyverbom, 2019; Kaun, 2020; Noble, 2018). However, there is sparse knowledge available on the occupational challenges and ethical risks that AICTs pose for professional communicators – the people employed to manage and perform the strategizing of communication. Given the novelty of these developments, the lack of research is not surprising, but attending to it is all the more urgent, as professional communicators’ new AICT-assisted strategy-practices are fast becoming organizational routines with significant social, economic, and political impact.
The research project Strategizing Communication and Artificial Intelligence (SCAI) addresses this knowledge gap by asking:
How are AI systems influencing professional communicators’ strategy-practices?
By tackling this specific issue, SCAI contributes with innovative research concerning a fundamental theme within the social sciences and humanities; namely, how to conceptualize the relationship between humans and machines. In particular, how intelligent computing affect human agency in relation to professional discretion and ethical judgement. Here, professional communicators constitute a ‘most likely critical case’ (Flyvbjerg, 2006) because changes in communication technologies affect their profession first and most deeply, as has been the case historically when e.g., the invention of the printing press led to the development of professional journalism and the rise of mass production led to the development of marketing (Peters, 1999). Following this logic, professional communicators are most likely to be influenced by artificially intelligent communication technology and studying their strategy-practices hence provide insights that are relevant to other professional settings as well as to people’s non-professional communicative practices.
Theoretical framework and methodological approach
SCAI takes its theoretical point of departure in the strategy-as-practice literature (Jarzabkowski & Spee, 2009; Kornberger & Clegg, 2011; Whittington, 2014), which integrates practice theory (Reckwitz, 2002), structuration theory (Giddens, 1984) and science and technology studies (MacKenzie & Wajcman, 1999). The strategy-as-practice perspective foregrounds how strategy is something organizations do – they strategize – and how that doing involves both human and non-human actors (e.g., AICT).
The approach has been further developed by communication scholars (Cooren, 2020) who apply a “processual understanding of the ‘practice turn’” (Mackay et al., 2020: 26) in order to forward the study of processes of strategizing communication (Gulbrandsen, 2019; Gulbrandsen & Just, 2020; King, 2009). Such processes of strategizing communication are conditioned by decision-making processes involving both human and non-human actors: they are constituted by the relations of a network of actors (assemblage), which are enabled and constrained by possibilities for action (affordances) and realized in concrete instances of action (agency) (Gulbrandsen & Just, 2016).
For the SCAI project, this means that when investigating how AICTs influence professional communicators’ strategy-practices, we must account for not only the outcome, but also the agential relations and particular actions of those involved in creating that outcome (Seaver, 2017). We must pay attention to the macro, meso and micro processes of strategizing (Whittington, 2006): At the macro level, we need to consider which actors (both human and non-human) are involved in the design, development, operationalization and governance of AICTs; at the meso level, we need to consider which potentials for design, development, operationalization and governance the combination of those actors afford; and at the micro level, we need to consider which concrete actions related to the development, design, operationalization and governance of AICTs those actors actually realise.
The framework affords an apt approach to studying the effects of artificially intelligent communication technologies as “contingent on the in-betweenness of a plethora of actors, both human and non-human” (Roberge & Seyfert, 2016: 2) and serve as the foundational principles for the project’s development of a theory of strategizing AICT.
Methodologically, the project is informed by organizational ethnography (Ybema et al., 2009), as this allows us to generate and collect data on the fine-grained sociomaterial interactions that constitute strategy-practices. By being ‘at the scene’, the project gains a better understanding of informants’ lived realities and how these realities connect to ‘panoramic processes’ (Hernes, 2014) – that is, we gain access to a multivocality of experiences and interpretations, where tensions and discrepancies can surface, and often-concealed dimensions of power and emotions become visible. With this approach, SCAI seeks to heed the strategy-as-practice literature’s call for attention to how micro-actions and macro-structures entangle in strategizing (Johnson et al., 2007). By taking a ‘long presence view’ (Kim et al., 2019), the project investigates how the influence of AICTs on professional communicators’ strategy-practices emerges and is established over time, an approach that resonates with recent research on strategic coherence (Lusiani & Langley, 2019).