In the latest episode of the Austrian Artificial Intelligence Podcast, I had the pleasure of speaking with Professor Phanish Puranam (INSEAD), a leading thinker on strategy and organizational design, and this year’s recipient of the Oscar Morgenstern Medal from the University of Vienna. Our conversation explored how technology—especially artificial intelligence—is reshaping organizations in subtle but powerful ways.
We began by looking at the question of centralization versus decentralization, a theme that runs through centuries of organizational history. As Professor Puranam explained, technologies themselves are neutral—they are “affordances” that can be adopted in different ways. The same tool can either empower employees with greater autonomy or provide managers with more powerful ways to monitor and control.
A simple example: collaboration platforms like Slack or Asana. Used one way, they allow teams to coordinate peer-to-peer, reducing the need for managerial oversight and thus decentralizing decision-making. Used another way, they can become instruments of constant surveillance, centralizing power in the hands of managers who track every communication. Technology does not dictate the outcome—design choices do.
Why Centralization Matters
Centralization offers speed and coordination. When rapid decisions are required, authority in the hands of a few can align action quickly. But decentralization offers diversity of thought, innovation, and—crucially—autonomy, which is a powerful motivator for human beings. The challenge, as Puranam notes, is always finding the right balance: harnessing the efficiency of centralized authority without eroding the creativity and intrinsic motivation that thrive in more decentralized settings.
Finding the right balance between efficient centralized control and the creativity and intrinsic motivation of decentralization.
Enter Generative AI
The second half of our conversation focused on Generative AI (GenAI) and its implications for work and skills. Here, the stakes are especially high. Unlike earlier predictive AI, which functioned as a tool, GenAI is crossing into the role of a “teammate.” It can draft documents, write code, generate analyses, and even perform managerial tasks such as evaluations or resource allocation.
But this power comes with risks. One is cognitive offloading: the tendency to outsource thinking to machines. Just as calculators can erode mental arithmetic skills, over-reliance on GenAI could hollow out human expertise over time. While AI might deliver short-term efficiency, organizations that allow skills to decay may find themselves at a long-term disadvantage.
There is also the question of differentiation. If every company has access to the same AI systems, what separates one organization from another? As Puranam puts it, the true competitive edge will not come from the algorithms—they are commodities—but from how humans use them. The ability to combine AI with human judgment, creativity, and organizational culture will define winners and losers.
Using GenAI without performing cognitive offloading that undermines the long term success of an organization.
The Human-Centric Organization
At the heart of the discussion lies a central idea: people still matter, perhaps more than ever. Research in psychology has shown that employees are motivated not only by pay, but by three key drivers: autonomy, relatedness, and competence. These are the very aspects most at risk when AI is deployed without care.
If companies suppress autonomy by centralizing control through AI, or undermine competence by letting machines perform all the thinking, they may find themselves struggling to retain talent. As Puranam argues, organizations must treat the design of AI-enabled work as an investment in their workforce. Allowing employees to co-develop with AI—reviewing its output, maintaining engagement, and continuing to build skills—may be slower and more costly in the short run, but it preserves the long-term vitality of the talent pipeline.
This is not just a “nice-to-have.” In a future where AI is ubiquitous, the only sustainable source of competitive advantage will be the unique capabilities of human workers—those who can learn, adapt, and create in ways that machines cannot.
Investing today, into human-centric AI that enhances employees capabilities will provide key competitive advantages in the future.
Echoes of Adam Smith
This theme resonates with an idea that dates back to Adam Smith’s Wealth of Nations. Smith argued that the prosperity of nations arises not from raw resources, but from the productive powers of labour, magnified by tools and machinery. In other words, technology mattered only insofar as people knew how to use it productively.
Today, AI is the machinery of our age. But simply deploying AI everywhere will not make an organization successful. The differentiator is having the right people—competent, motivated, and empowered—who know how to leverage these tools in distinctive ways. The “wealth of organizations” in the 21st century will come not from access to AI, but from how human talent collaborates with it.
AI can only succeed with a competent, motivated and empowered workforce.
Looking Ahead
As AI continues to evolve, the challenge for leaders will not be whether to adopt it—that is inevitable—but how to adopt it. Will it be used to flatten hierarchies while centralizing control in the hands of a few, or to empower employees with greater autonomy? Will it erode human skills, or be designed to enhance them?
Professor Puranam’s call is clear: resist the temptation to automate everything. Instead, invest in building human-centric organizations that treat AI not just as a tool for output, but as a catalyst for developing human capability.
Because in the end, it’s not the machines that will set organizations apart—it’s the people who know how to make those machines matter.
References
- Phanish Puranam – https://oskar-morgenstern-medaille.univie.ac.at/preistraegerinnen/phanish-puranam/
- Re-Humanize – https://www.penguin.sg/book/re-humanize/
- Scaling nonhierarchically: A theory of conflict‐free organizational growth with limited hierarchical growth – https://scholar.google.com/citations?view_op=view_citation&hl=en&user=T-5Hs8EAAAAJ&cstart=20&pagesize=80&sortby=pubdate&citation_for_view=T-5Hs8EAAAAJ:Aul-kAQHnToC
- Digital Collaboration Technologies and Managerial Intensity in US Corporations: An Examination – https://scholar.google.com/citations?view_op=view_citation&hl=en&user=T-5Hs8EAAAAJ&cstart=20&pagesize=80&sortby=pubdate&citation_for_view=T-5Hs8EAAAAJ:5icHVeHT4IsC
- Organizational cultural strength as the negative cross-entropy of mindshare: a measure based on descriptive text – https://www.nature.com/articles/s41599-022-01152-1
Are you unsure how to apply the learning of this episode in your organization, or want to dive deeper into the topic, we are here to help at Pasieka AI Solutions.
Thanks for reading. If you found this helpful, or have comments feel free to write an email at [email protected]