Current AI agents operate in isolation without high-level protocols for collaboration. This creates a critical gap for an 'internet of cognition,' which would enable agents to share context, understand intent, establish trust, and collectively solve problems, moving beyond siloed, human-mediated outputs.
The AI industry has focused on 'vertical scaling'—building bigger models with more parameters. Vijoy Pandey argues the untapped opportunity is in 'horizontal scaling.' This involves enabling teams of specialized agents to collaborate, creating a collective intelligence greater than any single model.
The classic 7-layer OSI networking model is insufficient for AI agents, which are non-deterministic endpoints. Cisco proposes adding Layer 8 for semantics (grammar) and Layer 9 for cognition (intent). These new layers would structure agent communication, enabling trust and governance.
Cisco internally developed CAPE, a multi-agent system of 20 distinct agents that manage complex cloud environments. This system has successfully automated 40% of tasks for site reliability engineers, reducing team load by 30% and cutting incident response times from hours to instantaneous.
Enterprises will not adopt multi-agent AI without two non-negotiable conditions. First, effective guardrails must be in place to ensure safety and compliance. Second, systems must be interoperable, as enterprises will inevitably use agents from diverse vendors like Salesforce, Microsoft, and Google, not a single provider.
Standard Role-Based Access Control (RBAC) is inadequate for dynamic AI agents. Cisco advocates for 'T-back': Tool, Task, and Transaction-based access control. This model grants agents ephemeral, minimum-necessary privileges only for a specific action, significantly enhancing security in autonomous systems.
To prevent a single company from controlling agent discovery and reputation like an app store, Cisco's open-source 'Agency' project builds its agent directory and identity systems on decentralized hash tables (DHTs). This ensures an open, interoperable ecosystem where no single entity is the gatekeeper.
The current state of AI development parallels early human evolution. Just as the invention of language enabled a step-function change in human collaboration and intelligence, AI agents now require their own 'language'—a set of shared protocols—to move beyond individual tasks and unlock collective problem-solving.
Cisco's SVP Vijoy Pandey reframes the company's core identity as enabling horizontal 'scale-out' through distributed systems. This directly contrasts with the dominant AI trend of 'scaling up' by creating ever-larger, monolithic models, positioning Cisco to power a future of collaborative, distributed AI.
The most efficient form of AI-to-AI communication could bypass natural language entirely. A proposed 'latent space transfer protocol' would allow agents to exchange their entire internal state (like a KV cache), akin to a neural link. This is currently feasible with open-weight models and promises huge efficiency gains.