What question did this study set out to answer?

The aim is to review the evolution of autonomous software agents and assess current reliability and governance challenges.

August 7, 2026Open Access

Agentic AI Between Capability and Reliability: A Structured Review of Architectures, Design Patterns, Evaluation Practice, and Enterprise Deployment

View Full Paper

Ask AI

Bookmark

Authors

Structured review identifies challenges in reliability and governance of agentic AI systems, implying a need for improved standards.

The aim is to review the evolution of autonomous software agents and assess current reliability and governance challenges.
Analysis of peer-reviewed papers and vendor documents from 2022 to 2026
Development of a reference architecture for agentic systems
Cataloguing design patterns and subsystem interactions
Agents now solve 80% of tasks on curated benchmarks (up from 20%)
Task completion on live interfaces ranges from 7% to 50%
Forecasts predict up to $1 trillion in annual spending on agentic software by 2030 despite low current enterprise adoption

Indrajit Sarkar (2026) studied this question.

View Full Paper

Ask AI

Bookmark