TechnologyK Puspa11/12/2026
India, Nov 12: Dynatrace, the AI-powered observability platform, today released findings from its State of Observability 2026 report, examining how organisations are navigating the growing complexity of AI adoption and the operational challenges associated with scaling autonomous systems. Based on a global survey of 1,202 technology leaders, the report finds that while enterprises are accelerating their adoption of AI and agentic technologies, their ability to maintain visibility, trust and control over these systems is struggling to keep pace.
The findings highlight this challenge in India, where 53% of organisations identify the expansion of AI, large language models and agentic workloads as a leading source of technology complexity, an increase of nine percentage points from the previous year. Unpredictable AI behaviour is another significant challenge, cited by 50% of Indian organisations. As enterprises move AI initiatives from experimentation into production, the growing volume of data, integration requirements and infrastructure demands are adding further pressure on technology teams.
The findings point to an emerging gap between AI ambition and operational proof – the ability of organisations to understand, validate and demonstrate what their AI systems are doing, whether they are behaving as intended and the business value they deliver.
As organisations increasingly deploy AI agents across enterprise environments, understanding how these systems make decisions is becoming critical to ensuring their reliability and trustworthiness. However, the report reveals a significant gap between what organisations consider important and what they currently have in place.
Globally, while 43% of organisations say understanding how AI agents reach decisions is important, only 27% currently address visibility into AI agent decisions. This highlights the challenge of ensuring transparency and accountability as enterprises give AI systems greater autonomy.
The scale of adoption is already significant, with 87% of organisations globally using AI agents moderately or extensively in observability and IT operations. However, only 12% have achieved fully autonomous operations today, despite 44% targeting this level of autonomy within the next five years. Humans continue to verify an average of 66% of AI-powered decisions, underlining the importance of human oversight as organisations build confidence in autonomous systems.
The rapid expansion of AI initiatives is creating new demands on enterprise technology environments. As organisations deploy more AI applications, they must manage increasingly complex workloads, unpredictable model behaviour and growing volumes of telemetry.
In India, 53% of organisations identify the expansion of AI, LLM and agentic workloads as a leading source of technology complexity, up nine percentage points from the previous year. This is accompanied by broader challenges, with 51% citing technology environment complexity and 50% pointing to unpredictable AI behaviour.
The growing volume of data generated by these workloads is another concern, with 45% of Indian organisations identifying telemetry volume as a source of complexity, an increase of 10 percentage points. These findings highlight the need for organisations to strengthen their ability to correlate and analyse data across increasingly distributed and AI-driven environments.
The global findings reinforce this trend, with 49% of organisations now running between six and 25 AI projects, compared with 42% a year ago. Furthermore, 43% globally identify the expansion of AI, LLM and agentic workloads as a leading source of technology complexity.
As AI adoption scales, managing the volume, quality and cost of data required to monitor enterprise environments is emerging as a significant operational challenge. Organisations need to collect and correlate data across traditional IT infrastructure and AI workloads to maintain a comprehensive view of system performance and behaviour.
Indian organisations report particularly pronounced challenges in this area:
The global findings further illustrate the financial and infrastructure implications of scaling AI. 81% of organisations report increased compute and infrastructure spending due to AI workloads, while 78% report higher energy costs.Additionally, 77% say AI workloads have increased their need for visibility into cost allocation.
These findings underline why enterprises need to move beyond simply scaling AI deployments towards establishing greater visibility into the resources, costs and operational requirements associated with them.
Despite the growing complexity, Indian organisations are seeing tangible benefits from observability as they navigate increasingly dynamic technology environments.
Reliability and uptime are identified as an important outcome by 68% of Indian organisations, an increase of 16 percentage points, while 57% cite better-informed decision-making, up 11 percentage points. These findings reflect the growing importance of observability in helping technology teams maintain operational continuity and make data-driven decisions as enterprise environments become more complex.
Capacity planning is also emerging as an important area of AI-driven value, cited by 41% of Indian organisations, an increase of 12 percentage points. However, the adoption of AI for broader operational efficiencies remains uneven. Only 28% identify security remediation and 25% cite toil reduction as key areas of AI value, indicating an opportunity for organisations to translate their AI investments into wider operational improvements.
As enterprises expand their use of AI and increasingly interconnected technology environments, maintaining security visibility and responding to vulnerabilities in a timely manner remain important operational priorities.
The report finds that 29% of Indian organisations report fixing vulnerabilities in under 24 hours, while only 21% have a target service-level agreement (SLA) of under 24 hours. This highlights a gap between actual remediation performance and formally established organisational targets, reinforcing the importance of aligning security practices, operational visibility and governance frameworks.
As AI systems become more deeply embedded in enterprise operations, organisations will need to ensure that their security and remediation processes evolve alongside their adoption of autonomous technologies.