Dynatrace Report Shows Half of Agentic AI Projects Stuck in Proof‑Concept Phase
Key Points
- Around 50% of agentic AI initiatives remain in proof‑of‑concept or pilot phases.
- Primary deployment focus: IT operations & DevOps, software engineering, and customer support.
- Highest expected ROI: IT operations monitoring, cybersecurity, and data processing.
- Major barriers: security/privacy/compliance, scaling management, and skilled‑staff shortages.
- 69% of AI decisions are still verified by humans; 87% of agents require supervision.
- 74% of leaders plan to increase AI budgets despite current challenges.
- Dynatrace recommends redefining ROI, setting clear guardrails, and scaling deliberately.
A recent Dynatrace study reveals that roughly half of organizations' agentic AI initiatives remain in proof‑of‑concept or pilot stages. While companies plan to raise AI budgets, progress is hampered by security, privacy, compliance concerns, difficulty managing agents at scale, and a shortage of skilled staff. Deployment focus is strongest in IT operations, DevOps, software engineering, and customer support, yet the greatest expected returns are in IT operations monitoring, cybersecurity, and data processing. Leaders emphasize human‑machine collaboration and recommend redefining ROI, establishing clear guardrails, and scaling deliberately.
Agentic AI Adoption Slows at Early Stages
A new Dynatrace report indicates that about fifty percent of agentic artificial intelligence projects are still confined to proof‑of‑concept or pilot phases. This suggests many organizations are struggling to move beyond experimentation toward full‑scale implementation, despite a strong appetite for AI‑driven transformation.
Key Deployment Areas
The study identifies the primary domains where companies are experimenting with AI agents: IT operations and DevOps (approximately 72% of respondents), software engineering (around 56%), and customer support (about 51%). These areas reflect where enterprises see immediate operational benefits from autonomous agents.
Expected Return on Investment
While investment focus is on the three domains above, the highest anticipated returns are projected in IT operations and system monitoring (roughly 44% of respondents), cybersecurity (about 27%), and data processing and reporting (around 25%). This gap highlights a mismatch between where money is being spent and where the greatest financial impact is expected.
Barriers to Progress
Several obstacles are slowing the transition from pilot to production. Security, privacy, and compliance concerns are cited by more than half of the participants. Managing and monitoring AI agents at scale presents a challenge for 51% of respondents, and a shortage of skilled staff or adequate training affects 44%. Additionally, one in three organizations point to an unclear business case as a barrier.
Human Oversight Remains Central
Human involvement continues to play a critical role in AI decision‑making. Approximately 69% of agentic AI decisions are still verified by humans, and 87% of companies are building agents that require human supervision. Moreover, 23% of respondents prefer to rely solely on human‑supervised agents for critical tasks.
Future Budget Outlook
Despite the challenges, confidence in AI remains high. Seventy‑four percent of surveyed leaders expect to increase their agentic AI budgets in the coming year, indicating a commitment to overcoming current hurdles.
Recommendations for Successful Scaling
Dynatrace advises organizations to rethink traditional ROI metrics, establish clear guardrails for human‑machine collaboration, and adopt a deliberate, intent‑driven scaling approach rather than large, indiscriminate spending. By focusing on measurable outcomes and robust governance, companies can better translate AI pilots into tangible business value.