Written by Martin Wilderer
Last week, the annual CES took place. I did not perceive it as “the next AI hype cycle,” but rather as a turning point.
My hypothesis:
AI has left the app layer. It is becoming part of physical systems and therefore a board-level, risk, and valuation factor.
What we saw was less spectacular than in previous years, and that is precisely what makes it so relevant in my view.
Thesis 1: “Physical AI” Is No Longer a Future Topic – It Is Operational
Robotics, vehicles, machines, devices: AI is no longer being added on, but embedded.
No longer as a chatbot, but as a standalone system consisting of sensors, models, software, updates, and security.
My observation:
The real value is not created by “AI inside” but by systems that actually work – systems that are maintainable, secure, and reliable.
If this is true, then the key management question is no longer:
“Where can we apply AI?”
but rather:
“Which of our products will become AI-defined systems—and which ones deliberately will not?”
Thesis 2: The Biggest Gap Is Not Technology, It Is Governance
In addition to what was shown, what was missing was equally telling:
Hardly any convincing answers around security, privacy, and liability.
Many AI features felt more like marketing add-ons than well-thought-out product decisions. An implicit “we’ll figure it out later,” even though AI has already entered physical operation.
My hypothesis here:
AI without governance will shift from an innovation risk to a reputational risk faster than many expect. I believe this is dangerously underestimated, especially at the board level.
Thesis 3: Interoperability Remains the Blind Spot With Real Cost Implications
Yes, there has been progress (e.g., in smart homes).
But in industry, robotics, and automotive, the reality remains: fragmented stacks, customer-side integration effort, and platform dependencies.
My thesis:
Integration will come. For now, companies still have to manage it themselves but they cannot afford to wait.
Why This Matters for Founders, Board Members, and Investors
If my hypotheses hold true, the valuation logic is shifting:
It is no longer just about market and margin,
but about system capability: update cycles, security, data sovereignty, platform strategy.
Not just an “AI strategy,” but AI product responsibility.
For PE investors, this means:
AI is no longer an add-on in the equity story deck—it becomes part of operational due diligence.
Three Questions I Would Currently Ask Boards
- What does our end-to-end value creation look like when AI is fully included?
- Where do we consciously accept platform dependencies—and where do we not?
- Have we solved security, privacy, and compliance at the product level, or only legally?
Conclusion
CES 2026 was not a “wow moment,” but it showed how deeply AI is already embedded in execution.
Even if much is still rough, this is no longer just about features—it is about end-to-end process design, integration into value creation, and, of course, governance.
AI no longer determines innovation alone, it determines liability, valuation, and trust.