The 3 questions to answer to take AI from experimentation to impact
A survey of over 1,200 IT leaders reveals that 60% of businesses already use autonomous systems, and 90% of executives say their AI rollouts exceed expectations. To turn enthusiasm into outcomes, companies must address three key questions: Are employees and governance ready? Are AI tools accessible? Do employees have the necessary capabilities? This blog explores how secure, governed AI agents can be integrated into natural workflows.
3 questions to answer to take AI from experimentation to impact | Databricks Blog
Skip to main content
Companies are starting to see the potential of AI in their businesses.
Today, 60% of businesses are already using autonomous systems in operations, according to a survey of over 1,200 IT technology leaders from Economist Enterprise. A resounding 90% of executives say their AI rollouts are beating expectations. And already, 75% of companies say they reworked job titles to reflect AI.
Now, as enterprises look to turn this enthusiasm into outcomes, the focus should be on bringing AI to users in intuitive, seamless ways that help them further improve productivity and efficiency. To make the pivot, business leaders to consider three key questions:
Are my employees and governance ready?
Are the AI tools accessible?
Do employees have the capabilities they need?
In this blog, we’ll discuss how to meet users where they are with secure, governed AI agents.
Are my employees and governance ready?
With a natural language interface, SQL expertise is not a prerequisite to generating business intelligence and technical prowess is no longer a blocker to automation. Instead, workers across the enterprise are increasingly able to deploy AI in unique, business-critical ways without advanced knowledge.
But to discover the blockbuster use cases, they need the freedom to safely experiment — and an understanding of how to wield new AI tools to deliver impact. There’s a gap between excitement and enablement.
Secure platforms help bridge the divide by giving employees the ability to safely test AI agents in many different scenarios. Without the right guardrails, companies may be forced to restrict employee usage, slowing down AI adoption and dampening the impact. Despite the benefits, less than half of companies have a formal governance framework in place for autonomous workloads, according to the Economist Enterprise survey. That isn’t tenable, and companies will eventually need to close these oversight gaps or face the consequences.
“Governance is not about slowing things down,” said Karthik Iyer, Group Vice President and Transformation Leader for Merchandising Technology and AI at Albertsons Companies. “It is what makes this level of speed and scale viable in the first place.”
When companies can impose the same governance framework across every AI workload, confidence builds. And employees are free to access the AI capabilities they need and hone new skillsets without jeopardizing the security of the business.
Are the AI tools accessible?
Forcing users to open new applications or switch to different tabs to access AI interfaces adds friction and impacts adoption. AI agents need to be accessible to employees in their natural workflows — whether that’s an office worker in front of a computer, or a retail worker on the store floor. From a single chat interface available across devices, they should be able to access all the company’s important data — from CRMs to Google Docs — for real-time insights and uninterrupted automation.
For example, AI interfaces can be embedded directly into the intelligence dashboards marketing teams live in everyday. After glimpsing at the broad view of operations, they can immediately turn to the AI agent to go deeper and investigate, act, or imagine. Instead of just asking What’s behind this spike in traffic?, teams can jump right to: How can we replicate this again in the future?.
The same capabilities can be integrated with other commonly used enterprise applications. But to enable seamless access to these tools, agentic interfaces must be supported by features including automated identity management to keep users constantly logged in, as well as consistent governance and business logic across every engagement.
“AI works best when it seamlessly integrates into the flow of any person’s working day,” said Ashish Agrawal, Chief Information Officer at KONE.
Do employees have the capabilities they need to get the most of AI?
When in-house tools are too restrictive, employees circumvent internal guardrails and “shadow IT” grows, creating significant governance challenges. Users want insights, but they want to be able to act on them. Increasingly, they want the AI agents to challenge their thinking, direct them to next steps, and even take action on their behalf. Ultimately, engaging with the systems should feel like partnering with skilled colleagues. AI agents have to go beyond just answering questions and instead harness the power of the full data estate to provide contextually-accurate, actionable intelligence and automation that doesn’t stop-and-stall. It needs to be an AI-worker that can take action on your behalf.
Review the full Economist Enterprise report to find out how leading organizations are making sure they meet employees where they are, with AI technology that actually makes a difference.
Get the latest posts in your inbox
Subscribe to our blog and get the latest posts delivered to your inbox.
Sign up
View all blogs