Balancing Innovation And Risk in The AI Age

AI adoption is a good example of how organizations try to balance innovation and risk. You can look at it as a series of races. First, to build the infrastructure to support AI; second, to train models; and third, to implement the models into actual business processes. Then comes the race that really matters: to create a healthy ROI from AI investments.

But to even be in the running, companies must outrun risks like cybersecurity, technical debt, and runaway costs. That’s one of the findings of our recent “Innovation Race” survey of 1,500 CIOs and IT decision makers. Just over half of the senior IT leaders (51%) report that they spend more time and resources on “firefighting”—keeping current operations up and running—instead of innovating to grow and improve the business. And a whopping 98% say their organization’s infrastructure needs improvement to support both risk management and innovation going forward.

Cyber threats surfaced as the biggest worry. And the risk isn’t limited to bad actors. Even human error, like a simple configuration mistake, can be enough to cause a major incident. But the risk landscape extends far beyond just cybersecurity.

While our respondents were certainly concerned about cybersecurity, they also mentioned:

Disparate systems and environments. The most common issue is with older systems or simply incompatible systems that must be un-siloed.

Technology debt. The infrastructure AI requires is costly and leaders know the needs will be dynamic. Eighty-one percent of survey respondents foresee data loads that will outgrow their organization’s current data centers.

Rising cloud costs. Cloud sticker shock has led organizations to rethink how they use the cloud and whether a hybrid approach will better suit their needs while controlling costs.

Energy costs: AI is also causing sticker shock with electricity bills. The amount of energy required for AI may also threaten an organization’s sustainability goals.

To this list, we might add:

The talent gap. One of the best ways to get out of firefighting doom loops is to have ample staff on hand. But the IT talent gap is real and it has immediate impacts, including higher costs.

Data and data management. Training AI models requires lots of data. Some organizations are feeling like they can’t even get out of the starting gates in the innovation race because they don’t have ready access to crucial data that’s stuck in siloed systems.

Looking for a path forward? Put the following strategies into action:

Understand your risk landscape. The risk landscape extends far beyond cybercrime or even customer-facing IT in general. Savvy leaders realize that the best approach is to scour your organization for vulnerabilities with a skeptic’s eye, ensuring that there are no surprises.

Assess AI readiness. AI hype and wishful thinking can dictate unrealistic AI timelines. Again, the mantra should be: no surprises. Be clear-eyed about AI readiness and set achievable goals that acknowledge the limits and obstacles specific to your organization.

Rethink infrastructure. Assess your needs, including AI goals, based on the specifics of your organization. Acknowledge that off-the-shelf solutions may only take you so far, and seek out the tools and partners that can help build the right systems.

Address cloud usage and cost: Ballooning cloud costs are driving many organizations to look at hybrid clouds. Seek out partners with favorable costs, and just as important, cost transparency.

Address energy costs. AI is clouding the sustainability picture for many organizations, in addition to the costs. Efficient, hybrid infrastructure with the right as-a-service partners can reduce energy usage, lower costs, and help shrink physical space requirements.

Establish a cyber recovery plan. Establish a robust recovery plan with multi-tiered resiliency, immutable and indelible snapshots, and ultrafast data recovery. Define roles, command hierarchies, and communication plans to use in case of a damaging cyber incident.

Bolster talent retention. If you have skilled people, they’re worth their weight in gold. Establish open communication and other processes to ensure that they’re happy with their roles, workloads, and career trajectory.

With these strategies you can give yourself a better chance to win your own innovation race.

Learn more about what your peers think about balancing risk and innovation in “The Innovation Race” survey.