Artificial intelligence ecosystems for marketing communications

The goal of this article is to help advertising scholars, students and practitioners understand and anticipate the effects of artificial intelligence (AI) and machine learning (ML) on advertising and, more generally, marketing communications (Marcom). While many discussions of AI centre on algorithms and models, we argue that to understand AI in Marcom, one must consider the broader ecosystem in which these algorithms operate. This article develops a framework that shows the Marcom-AI ecosystem and its outcomes, consisting of the following mutually reinforcing components: (1) algorithms and models, (2) customer data (3) digital environments (e.g. mobile devices, digital signage), (4) digital content assets (e.g. images, videos, copy) and (5) information technology infrastructure. We briefly sketch the uses of AI within Marcom. Most or all components of the ecosystem are usually necessary for AI to address Marcom opportunities and challenges. In conjunction with these components, the ecosystem comprises a broad set of stakeholders: consumers, influencers, brands/advertisers, media and messaging platforms, data platforms, publishers and content creators, MarTech/AdTech vendors, AI/ML service providers, device manufacturers and regulators. The combination of these components and stakeholders enables marketers to optimize touchpoints through targeting and choice architectures, create platforms for testing, derive insights from data, and support marketing processes and workflows. Building from the framework, we close by identifying future research directions for advertising scholars, including understanding consumer response to AI touchpoints, privacy, interactions between stakeholders, and how the ecosystem will evolve.