The digital twin is used in several problem domains. It is an enabler for Internet of Things applications as a framework for representing and simulating how physical objects interact in target environments. The concept of general-purpose sensing is aiming at determining basic sensing capabilities from which deriving, by means of artificial intelligence algorithms (termed synthetic sensing), relevant information for representing an environment. This article explores the possible relationships and the feasibility of an integration of synthetic sensing within a digital twin framework. This article presents relevant concepts and technologies, challenges and some enabled scenarios that this integration can bring. This approach is in its infancy and there is a need to validate and assess its viability, and benefits. This article identifies challenges and validation steps that can lead to consolidation and adoption of this approach. Finally, this article presents some future work aiming at demonstrating the approach.