This discussion, moderated by Rob van Kranenburg and Gaelle Le Gars, explores the transition from the Internet of Things (IoT) to an “Internet of Agents” (IoA).
Started with this LinkedIn post
Discussion with Vint Cerf, Elena Japundžić, Rob Tiffany, Gérald Santucci, Sophie Stalla-Bourdillon and moderated by Gaelle Le Gars and Rob van Kranenburg
Video is here
https://lnkd.in/ePsRb-3E
As we moved from the Internet to the Internet of Things we encountered architecture and governance issues regarding privacy and security that are yet to be resolved.
Asked Gemini to summarize
The core premise is that while the IoT connected objects to the network, the Internet of Agents involves autonomous or semi-autonomous AI entities that can act, negotiate, and make decisions on behalf of humans or systems.
Here is a summary of the key themes and perspectives contributed by the panelists:
1. The Architectural Shift (Vint Cerf)
Vint Cerf (Internet pioneer and VP at Google) emphasizes the technical and safety challenges of moving to an agent-based model.
- Interoperability: For an Internet of Agents to work, agents from different manufacturers must be able to communicate and “understand” each other’s capabilities and limitations.
- Safety & Responsibility: Cerf raises critical questions about accountability. If an agent makes a mistake (e.g., in a financial transaction or a smart home safety decision), who is responsible? The programmer, the owner, or the agent itself?
2. Digital Sovereignty and Governance (Gérald Santucci)
Gérald Santucci (former Director at the European Commission) focuses on the policy and governance framework.
- Privacy by Design: As we move toward agents that know our preferences intimately, the risk of surveillance increases.
- Strategic Autonomy: He discusses the need for a governance model that ensures these agents serve the public good and human rights, rather than just the interests of large tech corporations.
3. The Industrial Perspective (Rob Tiffany)
Rob Tiffany (expert in IoT and Digital Twins) brings the focus to the scale of deployment.
- Efficiency: Agents are seen as the “brain” of Digital Twins. In an industrial context, agents can manage supply chains or energy grids far more efficiently than static IoT sensors.
- Edge Computing: He notes that for agents to be effective, intelligence must be distributed at the “edge” (local devices) rather than relying solely on the cloud.
4. Legal and Ethical Frameworks (Sophie Stalla-Bourdillon)
Sophie Stalla-Bourdillon (Legal Counsel and Professor) addresses the “Legal Personality” of agents.
- The Social Contract: She explores how we define the relationship between humans and autonomous agents.
- Data Protection: She argues that current GDPR frameworks may be insufficient for a world where agents are constantly interacting and sharing data autonomously.
5. Design and Human Experience (Elena Japundžić)
Elena Japundžić (Designer and Researcher) discusses the “Internet of People” within this machine-driven world.
- Agency vs. Automation: There is a tension between making life easier through automation and stripping humans of their agency.
- Trust: The success of the Internet of Agents depends on whether humans can trust these “black box” algorithms to act in their best interests.
Core Conclusion
The panel concludes that the Internet of Agents is not just a technological upgrade, but a paradigm shift. We are moving from a web of information (Internet) to a web of things (IoT), and finally to a web of action (IoA). This requires new standards for interoperability, updated legal frameworks for liability, and a renewed focus on ensuring that “intelligence” remains under human-centric control.
create a WordPress feature graphic about this Summary

Google Privacy PolicyOpens in a new windowGoogle Terms of ServiceOpens in a new windowYour privacy & Gemini AppsOpens in a new window
Gemini may display inaccurate info, including about people, so double-check its responses.
Internet of Agents: A Paradigm Shift
