The advent of newly available AI agents promises to take app modernisation to new heights, a welcome addition among developers who work in fast-paced and dynamic software development environments.
“AI agents represent the most promising technology available to enterprise developers since intelligent process automation (IPA) began addressing cumbersome digital transformation deployments in past years,” said Charlotte Dunlap, Research Director for GlobalData.
“Despite the strides made in platform services including automation, CICD pipelines, and Kubernetes open-source technology, enterprises continue to struggle with defining, creating, and managing business transformations in a way that modernises traditional infrastructure and applications.”
“The difference is that while IPAs automate repetitive, rule-based business processes, AI agents support goal-driven applications capable of adapting and reasoning in an autonomous fashion,” she added.
Increasing implementation of AI agents
As a result, organisations have begun pivoting from building deterministic workflows to implementing AI agents that learn, adapt, make decisions and perform complex tasks using reasoning and longer-term memory, acting independently.
With these changes comes the need to better define the market for AI agents. GlobalData has identified key components which enterprises will want to take into consideration as they compare competing and comprehensive AI agent portfolios:
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalData- Mature ML and natural language processing (NLP) technologies, foundational to well-known AI assistants, now serving as the underpinnings of agentic AI offerings
- At the heart of agentic innovations are implementation frameworks and libraries capable of perception, reasoning, memory, planning, and action
- Standardised protocols (e.g., MCP, A2A, ACP, etc.) support interoperability
- Agent orchestration for managing agents across the enterprise to ensure data quality and governance
- Low-code development tools/interface to democratise build/deployment of AI agents across the enterprise
- Integration capabilities (and/or APIs) between AI-injected applications and systems to streamline new workflow automation
- Unified data management with governance and monitoring features supported by an attractive UX
Standardised protocols in particular have only just begun to gain attention for their support of better communication and collaboration between multi-agent systems.
“This type of improved integration will help drive agentic technology forward. The standards also play a vital role in helping to democratise AI agents among members of the DevOps team by making agents more accessible to broader personas, not savvy in infrastructure configuration,” Dunlap added.
Why automation works
Despite significant technology advancements in platform services over the years, including low-code interfaces, intelligent automation, and microservices containerisation, business transformations are often considered costly, slow, and laborious. This hardship inevitably results in a lack of executive support, internal expertise, and access to adequate consultants and integrators.
Agentic AI promises even greater automation to a range of complex tasks transforming business processes. They are designed to solve problems, attack tasks with minimal human input, and learn and become even more effective over time.
