An autonomous agent is a complete software system, not merely a feature of an LLM. Dell's CTO defines it by four key components: an LLM (for reasoning), a knowledge graph (for specialized memory), MCP (for tool use), and A2A protocols (for agent collaboration).
Dell's CTO warns against "agent washing," where companies incorrectly label tools like sophisticated chatbots as "agentic." This creates confusion, as true agentic AI operates autonomously without requiring a human prompt for every action.
When multiple AI agents work as an ensemble, they can collectively suppress hallucinations. By referencing a shared knowledge graph as ground truth, the group can form a consensus, effectively ignoring the inaccurate output from one member and improving overall reliability.
If applying GenAI to a process doesn't improve key metrics like revenue or cost, it's a sign that the original human task was likely low-value or "BS work." The AI exposes work that doesn't contribute to business outcomes, prompting re-evaluation of its necessity.
The "AI ROI flywheel" is a strategy where an organization starts with AI projects that deliver massive, measurable returns (e.g., 10:1 to 30:1). These initial wins create credibility and buy-in, making it progressively easier to secure resources for future AI initiatives.
To avoid "AI slop"—the proliferation of low-quality AI outputs—Dell's CTO advocates for a disciplined, top-down strategy. Instead of letting tools run wild, they focus on a small number of high-impact use cases with clear business outcomes, ensuring quality and preventing chaos.
Dell's CTO acknowledges the Model Context Protocol (MCP) is powerful for agent tool access but isn't yet enterprise-grade. To manage this risk, Dell centralizes all its MCP servers into a single controlled environment, allowing them to wrap the immature protocol with robust security controls.
Dell's CTO, Jon Rose, reveals a historic achievement: decoupling revenue growth from cost increases. By focusing on high-ROI AI projects in key business areas, they grew revenue by $10 billion while simultaneously reducing absolute costs, a first in the company's 41-year history.
Dell's CTO identifies a new architectural component: the "knowledge layer" (vector DBs, knowledge graphs). Unlike traditional data architectures, this layer should be placed near the dynamic AI compute (e.g., on an edge device) rather than the static primary data, as it's perpetually hot and used in real-time.