Skip to main content

AI Innovation Was the First Phase. Governance and Accountability Are the Next ...

Bhaskar Challa
Techwave

The race to deploy AI is largely over. Most enterprises have entered it. The question now is not whether artificial intelligence is running inside the organization. The question is whether anyone is genuinely responsible for what it does. That is not a technical question. It is a leadership one. And most organizations are not yet structured to answer it honestly.

What Nobody Wants to Say Out Loud

There is a version of this conversation that gets sanitized in boardrooms and skipped in keynotes. The honest version goes something like this: Thousands of alerts are generated by enterprise systems every single day. Over half carry no meaningful signal. Teams across industries spend between 60% and 70% of their operational time managing noise rather than solving problems that matter. Critical failures sit buried in that volume, invisible until they are not. And somewhere in the middle of all this, organizations deployed AI workloads on top of infrastructure that was already struggling to be understood.

The cost is not abstract. A single business-critical outage in a digitally dependent sector can cost upward of $9,000 every minute it persists. Most large organizations experience at least one such event every year. And yet the response in most enterprises has been to add more tools rather than build better intelligence. This is the environment in which the accountability conversation has become unavoidable.

The Tools That Served Us Then Cannot Serve Us Now.

The monitoring infrastructure that governed IT operations a decade ago was designed for a world that no longer exists. Static rules. Predictable failure modes. Infrastructure that a reasonably sized team could hold in their collective awareness.

Modern enterprise technology does not work that way. Systems run across layered cloud environments, containerized workloads, microservices, and external integrations that multiply failure paths beyond what any ruleset can anticipate. Research puts the proportion of alerts either ignored or acted on too late at close to 40%. That is not a figure representing careless teams. It represents the structural impossibility of asking humans to govern complexity at a scale that exceeds what human attention can reliably cover.

The answer is intelligence embedded in the operations layer itself. Machine learning applied to IT operations gives organizations the ability to correlate signals across the entire estate, separate genuine anomalies from noise, and surface problems before they reach customers. Organizations with mature implementations are reporting incident resolution improvements of 30% to 50%. More significantly, research consistently shows that more than half of all outages could be prevented if early warning signals were detected and acted on in time. Prevention, not reaction. That distinction changes the nature of IT leadership entirely.

Accountability Is a Record, Not a Promise

Speed matters. But accountability is something different from speed. What accountability requires at an operational level is concrete. Live telemetry across every system the AI touches. Behavioral monitoring that detects when a system drifts from its intended parameters. Compliance verification that does not depend on a human noticing something unusual after the fact. All of it is embedded in the architecture from the beginning, not retrofitted after an incident has already occurred.

Governance built in and governance applied after the fact produce entirely different outcomes. One demonstrates through continuous evidence that systems are operating within their boundaries. The other produces documentation after something has gone wrong and calls it a response. Most organizations today sit closer to the second position than the first. That reflects how quickly AI deployment moved relative to how slowly governance infrastructure typically develops. The pressure to deploy was real. What did not keep pace was the operational infrastructure to govern what was being built. That gap is now closing, not because organizations chose to slow down, but because the consequences of leaving it open have become impossible to manage quietly.

The Regulatory Timeline Is Already Behind You

There is a version of the AI governance conversation framed as preparation. Build the framework now so you are ready when regulation arrives. That framing is already out of date.

Regulators across multiple jurisdictions are asking organizations to demonstrate, today, that their AI systems operate within defined ethical and operational boundaries, that those boundaries are monitored on an ongoing basis, and that a clear chain of responsibility exists when a system behaves outside them. These are not aspirational standards. They are active requirements in regulated sectors and emerging requirements in sectors that have not historically faced this level of scrutiny.

Every principle must map to a specific control. Every control must have a measurable check and a defined owner. The distance between a published AI ethics statement and that kind of operational infrastructure is the distance between intent and accountability. Closing it is the work of this period.

The People This Actually Affects

The human dimension of this challenge deserves to be said plainly. The engineers and platform teams inside organizations that lack adequate visibility are not failing at their jobs. They are doing demanding work in conditions that make it unnecessarily hard. Spending the majority of the day managing alerts that carry no signal, diagnosing failures that could have been anticipated, and restoring systems under pressure without a clear picture of what caused the problem. That is exhausting work. It produces burnout, not insight. And it wastes the capability that those teams actually possess.

When the intelligence layer improves, the nature of the work changes. Teams that were spending most of their energy reacting begin doing something genuinely different. They anticipate. They build.

As AI operations mature toward greater autonomy, where systems can monitor their own conditions and initiate corrective action without waiting for human instruction, the role of the people alongside those systems shifts in genuinely positive ways. Less time watching. More time thinking. That is not a threat to the people doing this work. It is a better version of the work.

The Shift We See Playing Out in Practice

Organizations are moving beyond isolated AI deployments toward integrated cloud operating models where governance, intelligence, and accountability are built into every layer. From assessment, architecture design to operations and optimization. This means embedding AIOps-driven monitoring to eliminate noise and detect anomalies early, implementing FinOps frameworks to enforce cost accountability, and establishing unified governance models that bring together security, compliance, and operational control. Enterprises that operationalize AI in this way gain continuous visibility across hybrid cloud environments, reduce incident response times through automation and self-healing systems, and create measurable accountability through defined KPIs, SLAs, and governance structures. Ultimately, the organizations that succeed will not be the ones that deploy AI fastest, but those that can govern it and be accountable by design.

Bhaskar Challa is Associate VP and Global Head for Cloud & Infra Managed Services at Techwave

The Latest

Automation often starts with the people closest to an immediate need. A developer may schedule a script with cron or Amazon EventBridge Scheduler; a business team may adopt an RPA bot, low-code platform, or agent. Many tools stay useful within one domain. Some become consequential across organizational boundaries because other systems and teams depend on them or business outcomes rely on their timing and recovery. Those cases may call for broader coordination ...

The conventional wisdom in our industry says that AI demands a new operating model and specific data tailored for it. I disagree. AI is changing the rules, but not in the way people think ...

Most organizations invest heavily in security tools, policies, and controls. Yet despite these efforts, one reality persists: employees often find ways around them. This isn't always malicious. In many cases, it’s a response to friction, urgency, or misalignment between how work actually happens and how systems are designed to protect it. When employees bypass security, it creates blind spots. Sensitive data may move through unmonitored channels, access controls may be weakened, and risks that security teams intended to mitigate reappear in new forms ...

Enterprise AI projects are often justified with a familiar metric: hours saved. A team automates a manual process, employees spend fewer hours reviewing documents or entering data, and the productivity gain becomes the headline. Time savings matter, but they capture only part of the financial impact AI can have inside an organization ... A more useful way to evaluate enterprise AI is to examine the financial decisions and risks surrounding the workflows it supports ...

We run an internal AI assistant that answers engineering questions in Slack ... Each question kicks off what we call a turn ... We could already audit a single turn after the fact. Every reply carries a short lookup token in its footer, and a slash command pulls up the full receipt for that turn. What we could not do was answer questions about the whole fleet. How long do turns take? What do they cost? Which tools does the agent reach for? Where does the time go inside a single answer? Those are APM questions. So we treated the agent like any other service and instrumented it with OpenTelemetry, end to end. This post covers what we built, the design choices that turned out to matter, and the things that surprised us ...

Applications have long used bounded, deterministic logic to control what happens within their own functions. Process orchestration coordinates activity across people, applications, decisions, and automation tools. Workload orchestration coordinates executable work and dependencies across applications, platforms, and infrastructure. As agentic capabilities enter applications and domain systems, more planning and decision-making will also occur inside those environments. These forms of automation will coexist. Sometimes they will overlap, and they will increasingly need to exchange triggers, state, and results. But they are not becoming interchangeable ...

For many organizations, the AI agenda has landed squarely on the CIO's desk. At first glance, that makes perfect sense. AI depends on data, infrastructure, security, integration, governance and scalable technology platforms — all areas where the CIO plays a central role. But there is a problem with treating AI primarily as an IT initiative: AI is not simply a technology deployment. It’s also a business transformation ...

In MEAN TIME TO INSIGHT Episode 28, Shamus McGillicuddy, EMA VP of Research, Network Infrastructure and Operations, and special guest Sanjay Gidwani, Founder & CEO of Kosmos AI Labs, discuss root cause analysis ... 

A gap is growing between how quickly organizations are putting AI to work and how much autonomy they are prepared to give it, according to The 2026 State of Observability Report from Grafana Labs ...

AI agents and agentic browsers now make up a growing share of the traffic reaching websites and applications. They search, compare, sign up, and increasingly complete purchases on behalf of the people who delegate those tasks. Digital experience monitoring (DEM) was built on the assumption that every session is performed by a person. That assumption no longer holds, and the measurements built on it are shifting without notice ...

AI Innovation Was the First Phase. Governance and Accountability Are the Next ...

Bhaskar Challa
Techwave

The race to deploy AI is largely over. Most enterprises have entered it. The question now is not whether artificial intelligence is running inside the organization. The question is whether anyone is genuinely responsible for what it does. That is not a technical question. It is a leadership one. And most organizations are not yet structured to answer it honestly.

What Nobody Wants to Say Out Loud

There is a version of this conversation that gets sanitized in boardrooms and skipped in keynotes. The honest version goes something like this: Thousands of alerts are generated by enterprise systems every single day. Over half carry no meaningful signal. Teams across industries spend between 60% and 70% of their operational time managing noise rather than solving problems that matter. Critical failures sit buried in that volume, invisible until they are not. And somewhere in the middle of all this, organizations deployed AI workloads on top of infrastructure that was already struggling to be understood.

The cost is not abstract. A single business-critical outage in a digitally dependent sector can cost upward of $9,000 every minute it persists. Most large organizations experience at least one such event every year. And yet the response in most enterprises has been to add more tools rather than build better intelligence. This is the environment in which the accountability conversation has become unavoidable.

The Tools That Served Us Then Cannot Serve Us Now.

The monitoring infrastructure that governed IT operations a decade ago was designed for a world that no longer exists. Static rules. Predictable failure modes. Infrastructure that a reasonably sized team could hold in their collective awareness.

Modern enterprise technology does not work that way. Systems run across layered cloud environments, containerized workloads, microservices, and external integrations that multiply failure paths beyond what any ruleset can anticipate. Research puts the proportion of alerts either ignored or acted on too late at close to 40%. That is not a figure representing careless teams. It represents the structural impossibility of asking humans to govern complexity at a scale that exceeds what human attention can reliably cover.

The answer is intelligence embedded in the operations layer itself. Machine learning applied to IT operations gives organizations the ability to correlate signals across the entire estate, separate genuine anomalies from noise, and surface problems before they reach customers. Organizations with mature implementations are reporting incident resolution improvements of 30% to 50%. More significantly, research consistently shows that more than half of all outages could be prevented if early warning signals were detected and acted on in time. Prevention, not reaction. That distinction changes the nature of IT leadership entirely.

Accountability Is a Record, Not a Promise

Speed matters. But accountability is something different from speed. What accountability requires at an operational level is concrete. Live telemetry across every system the AI touches. Behavioral monitoring that detects when a system drifts from its intended parameters. Compliance verification that does not depend on a human noticing something unusual after the fact. All of it is embedded in the architecture from the beginning, not retrofitted after an incident has already occurred.

Governance built in and governance applied after the fact produce entirely different outcomes. One demonstrates through continuous evidence that systems are operating within their boundaries. The other produces documentation after something has gone wrong and calls it a response. Most organizations today sit closer to the second position than the first. That reflects how quickly AI deployment moved relative to how slowly governance infrastructure typically develops. The pressure to deploy was real. What did not keep pace was the operational infrastructure to govern what was being built. That gap is now closing, not because organizations chose to slow down, but because the consequences of leaving it open have become impossible to manage quietly.

The Regulatory Timeline Is Already Behind You

There is a version of the AI governance conversation framed as preparation. Build the framework now so you are ready when regulation arrives. That framing is already out of date.

Regulators across multiple jurisdictions are asking organizations to demonstrate, today, that their AI systems operate within defined ethical and operational boundaries, that those boundaries are monitored on an ongoing basis, and that a clear chain of responsibility exists when a system behaves outside them. These are not aspirational standards. They are active requirements in regulated sectors and emerging requirements in sectors that have not historically faced this level of scrutiny.

Every principle must map to a specific control. Every control must have a measurable check and a defined owner. The distance between a published AI ethics statement and that kind of operational infrastructure is the distance between intent and accountability. Closing it is the work of this period.

The People This Actually Affects

The human dimension of this challenge deserves to be said plainly. The engineers and platform teams inside organizations that lack adequate visibility are not failing at their jobs. They are doing demanding work in conditions that make it unnecessarily hard. Spending the majority of the day managing alerts that carry no signal, diagnosing failures that could have been anticipated, and restoring systems under pressure without a clear picture of what caused the problem. That is exhausting work. It produces burnout, not insight. And it wastes the capability that those teams actually possess.

When the intelligence layer improves, the nature of the work changes. Teams that were spending most of their energy reacting begin doing something genuinely different. They anticipate. They build.

As AI operations mature toward greater autonomy, where systems can monitor their own conditions and initiate corrective action without waiting for human instruction, the role of the people alongside those systems shifts in genuinely positive ways. Less time watching. More time thinking. That is not a threat to the people doing this work. It is a better version of the work.

The Shift We See Playing Out in Practice

Organizations are moving beyond isolated AI deployments toward integrated cloud operating models where governance, intelligence, and accountability are built into every layer. From assessment, architecture design to operations and optimization. This means embedding AIOps-driven monitoring to eliminate noise and detect anomalies early, implementing FinOps frameworks to enforce cost accountability, and establishing unified governance models that bring together security, compliance, and operational control. Enterprises that operationalize AI in this way gain continuous visibility across hybrid cloud environments, reduce incident response times through automation and self-healing systems, and create measurable accountability through defined KPIs, SLAs, and governance structures. Ultimately, the organizations that succeed will not be the ones that deploy AI fastest, but those that can govern it and be accountable by design.

Bhaskar Challa is Associate VP and Global Head for Cloud & Infra Managed Services at Techwave

The Latest

Automation often starts with the people closest to an immediate need. A developer may schedule a script with cron or Amazon EventBridge Scheduler; a business team may adopt an RPA bot, low-code platform, or agent. Many tools stay useful within one domain. Some become consequential across organizational boundaries because other systems and teams depend on them or business outcomes rely on their timing and recovery. Those cases may call for broader coordination ...

The conventional wisdom in our industry says that AI demands a new operating model and specific data tailored for it. I disagree. AI is changing the rules, but not in the way people think ...

Most organizations invest heavily in security tools, policies, and controls. Yet despite these efforts, one reality persists: employees often find ways around them. This isn't always malicious. In many cases, it’s a response to friction, urgency, or misalignment between how work actually happens and how systems are designed to protect it. When employees bypass security, it creates blind spots. Sensitive data may move through unmonitored channels, access controls may be weakened, and risks that security teams intended to mitigate reappear in new forms ...

Enterprise AI projects are often justified with a familiar metric: hours saved. A team automates a manual process, employees spend fewer hours reviewing documents or entering data, and the productivity gain becomes the headline. Time savings matter, but they capture only part of the financial impact AI can have inside an organization ... A more useful way to evaluate enterprise AI is to examine the financial decisions and risks surrounding the workflows it supports ...

We run an internal AI assistant that answers engineering questions in Slack ... Each question kicks off what we call a turn ... We could already audit a single turn after the fact. Every reply carries a short lookup token in its footer, and a slash command pulls up the full receipt for that turn. What we could not do was answer questions about the whole fleet. How long do turns take? What do they cost? Which tools does the agent reach for? Where does the time go inside a single answer? Those are APM questions. So we treated the agent like any other service and instrumented it with OpenTelemetry, end to end. This post covers what we built, the design choices that turned out to matter, and the things that surprised us ...

Applications have long used bounded, deterministic logic to control what happens within their own functions. Process orchestration coordinates activity across people, applications, decisions, and automation tools. Workload orchestration coordinates executable work and dependencies across applications, platforms, and infrastructure. As agentic capabilities enter applications and domain systems, more planning and decision-making will also occur inside those environments. These forms of automation will coexist. Sometimes they will overlap, and they will increasingly need to exchange triggers, state, and results. But they are not becoming interchangeable ...

For many organizations, the AI agenda has landed squarely on the CIO's desk. At first glance, that makes perfect sense. AI depends on data, infrastructure, security, integration, governance and scalable technology platforms — all areas where the CIO plays a central role. But there is a problem with treating AI primarily as an IT initiative: AI is not simply a technology deployment. It’s also a business transformation ...

In MEAN TIME TO INSIGHT Episode 28, Shamus McGillicuddy, EMA VP of Research, Network Infrastructure and Operations, and special guest Sanjay Gidwani, Founder & CEO of Kosmos AI Labs, discuss root cause analysis ... 

A gap is growing between how quickly organizations are putting AI to work and how much autonomy they are prepared to give it, according to The 2026 State of Observability Report from Grafana Labs ...

AI agents and agentic browsers now make up a growing share of the traffic reaching websites and applications. They search, compare, sign up, and increasingly complete purchases on behalf of the people who delegate those tasks. Digital experience monitoring (DEM) was built on the assumption that every session is performed by a person. That assumption no longer holds, and the measurements built on it are shifting without notice ...