China’s elder care industry is hoping automation will provide the solution to an ageing population and a shrinking workforce – Copyright AFP HECTOR RETAMAL

In what way is AI is reshaping data skills, teams, and upskilling strategies in the new year? For a strategic insight, Digital Journal has interviewed Iris Adae, VP of Data & Analytics at KNIME. In answering our questions, Adae uses her personal expertise and observations in the industry.

This is in keeping with Digital Journal’s current theme of technological related predictions for 2026.

Digital Journal:  As AI and data tools become more accessible, how are organizations rethinking upskilling to balance automation with human oversight and domain expertise?

Iris Adae: Over the past few years, we’ve experienced one of the fastest periods of technological change we’ll likely see for a long time. Just 30 years ago, mobile phones were rare and mostly used by technical early adopters. Today, AI helps automate repetitive work, rearrange schedules, and retrieve information far faster than traditional search engines. This pace of change won’t slow down. Even though I believe we’ve reached peak hype right now, the hype will soon settle into regular, everyday usage.

The past year was about experimentation: trying things out, challenging models, and, importantly, failing. Now we need to take what we’ve learned and build reliable, reusable AI frameworks and usage patterns. That’s the challenge ahead.

A crucial part of this transition is upskilling non-data professionals so they can use AI and automate repetitive tasks themselves. There are three pillars that organizations need to prioritize to be ready:

  1. Basics: Every organization should be investing in foundational AI learning. What works best for us are short online courses that begin with data and AI literacy, since that knowledge is essential for using AI and automation effectively. From there, an AI usage and best practices course is extremely valuable, especially one that covers how prompt engineering works. Finally, a hands-on course focused on building an actual AI solution makes the learning stick, especially if employees can continue using the solution they build in their daily work.
  • Collaboration: Collaboration is the most effective way to teach and learn. My data team regularly partners with business teams to implement their use cases. Recently, we helped our finance team fully automate their cash reporting. The key is that the data team drives and moderates the project, but the finance team ultimately owns the automation, making changes and improvements independently.
  • Tools that enable transparency and human oversight:  Many organizations are investing in powerful AI tools for everything from support tickets to lead-to-cash automation. While tempting, I recommend choosing more transparent solutions that keep humans in the loop. Low-code and no-code platforms give users visibility into what the AI is doing, especially when paired with automation features.

For my own team, AI often serves as a filter. Instead of manually reviewing 100-page KPI reports, we have AI generate a pre-filtered set of insights. We evaluate those insights and typically send one or two forward as actionable recommendations. As we adopt AI more deeply, we must also increase awareness of the AI Act, which will continue rolling out in 2026. Leaders need to understand upcoming regulations.

DJ: What are the three highest-value skills data teams will need by 2026, and how will these roles evolve as automation becomes more widespread?

Adae: In recent years, we’ve learned which tasks AI can reliably support (especially repetitive or tedious work) and where human oversight remains essential. As we automate these tasks, we free up valuable resources.

The first major shift will be in Data Engineering. Clean data is absolutely essential for AI models. Because these models are limited by the data they’re trained on, poor data quality directly leads to flawed results. Organizations must prioritize investment in Data Engineering teams, both by giving them advanced tools and by increasing headcount. A second key area is Data Operations. These teams maintain AI tools and systems, and in some organizations, they may sit within the Data Engineering function. Third, companies should rethink how they allocate Business Analyst time. Instead of hiring more analysts, organizations should use AI to filter and pre-process information so analysts can focus entirely on human oversight and applying domain expertise.

Teams should also reassess their current projects and routines. Many tasks simply no longer need to exist. For example, a team recently discovered they were sending quarterly Excel reports that neither the sending team nor the receiving team actually needed. A quick conversation eliminated the entire process. The era of passively emailing reports is over. Data teams must shift into a proactive, insights-driven mode. This is the perfect time to clean up outdated projects and free resources for building new AI workflows.

DJ: How should organizations structure AI ownership as they prepare for deeper AI integration?

Adae: Over the past few years, some companies have created dedicated AI teams within their Data & Analytics functions, while others have distributed AI responsibilities across existing business teams. While both approaches can work, I strongly believe that one specific role is essential for everyone: a dedicated AI strategist who owns AI strategy and implementation across the company. This person must drive cross-team AI projects and lead the change management process required to make AI adoption successful. Without someone accountable for steering the organization’s AI mindset and roadmap, progress will stall.

DJ: If you had to summarize the most important priorities for companies preparing for AI in 2026, what would they be?

Adae: The most important priorities for the years ahead are clear: organizations must recognize that data quality matters far more than trying to enhance or refine their models, they need a dedicated AI strategist who is responsible for driving AI-related change across the company, and they should clean up outdated projects and workflows so teams can free up time and resources to transition into new, AI-based routines.

Source link