‍1) The key assumptions behind Huang's prediction:

• Nearly all  current IT systems and software applications are inherently deterministic and cannot support the unpredictable and probabilistic nature of Generative AI (GenAI).

Therefore, it is unsurprising that, according to a recent Gartner report, GenAI is currently most useful for content generation and conversational user interfaces, not yet for decision support, automation, business orchestration, etc. (Learn more from the Gartner report: https://gtnr.it/45s622W)This is because most automation software solutions, including workflow, RPA, BPM/DPA, and iPaaS, are also created upon an inherently deterministic architecture to support primarily predefined and structured automation and orchestration.

2) To build an AI Factory, organizations must ensure that existing IT systems and software applications can be reused and coexist with it.

• Organizations must be able to build applications that can support probabilistic use cases, i.e., GenAI-ready, on the existing deterministic-in-nature IT infrastructure.

AgilePoint, an abstraction-enabled codeless composable application platform, enables model-driven cross-platform composability. This results in applications being represented 'as a model' in metadata, free of code generation and compilation and without needing virtual machines, to be probabilistic-ready for utilizing GenAI.

At the CxO Summit, we demonstrated how organizations could reuse more than 110 popular IT and automation systems already supported by AgilePoint to create applications that can support unpredictable and probabilistic use cases and gain strategic advantages by utilizing GenAI.

Therefore, the keys to accelerating AI operationalization are:

1) Ensuring the building of future enterprise applications, automation, and orchestrations on a new generation of probabilistic-ready application platforms to future-proof for AI and GenAI readiness; and

2) Enabling the coexistence of 'Software Factory' and 'AI Factory.'