Abstractions for AI-Based User Interfaces and Systems
This addresses programming challenges for developers building AI-based interfaces like natural-language agents, though it appears incremental as foundational steps toward a complete language.
The paper tackles engineering challenges in AI-based user interfaces by proposing three programming language abstractions for intelligent system design: hypothetical worlds for nondeterministic search, a feature type system for application-learning algorithm interaction, and collaborative execution constructs for multi-machine scenarios with privacy control.
Novel user interfaces based on artificial intelligence, such as natural-language agents, present new categories of engineering challenges. These systems need to cope with uncertainty and ambiguity, interface with machine learning algorithms, and compose information from multiple users to make decisions. We propose to treat these challenges as language-design problems. We describe three programming language abstractions for three core problems in intelligent system design. First, hypothetical worlds support nondeterministic search over spaces of alternative actions. Second, a feature type system abstracts the interaction between applications and learning algorithms. Finally, constructs for collaborative execution extend hypothetical worlds across multiple machines while controlling access to private data. We envision these features as first steps toward a complete language for implementing AI-based interfaces and applications.