The
Future Of Enterprise AI

In this episode of “The Future Of,” Jeff Dance interviews Douwe Kiela, CEO and co-founder of Contextual.ai and co-inventor of the RAG (Retrieval Augmented Generation) technique. They discuss the evolution and future trajectory of enterprise AI, emphasizing the importance of context engineering over simple prompt engineering, and explore the challenges and opportunities associated with integrating language models in complex enterprise environments. The conversation covers current adoption trends, the value of multi-agent systems in AI, and the necessity for businesses to focus on both cost savings and new revenue generation. Ethical considerations and the broader societal impact of advanced AI technologies are also examined.

The Evolution of Enterprise AI

• From Demos to ROI: While 2023 was the “year of the demo” and 2024 focused on production use cases, 2025 is shifting toward proving actual Return on Investment (ROI).

• Context Engineering: Prompt engineering is no longer sufficient; enterprises now require a “context layer” that connects noisy, messy data sources (SharePoint, Google Drive, SQL) to language models.

• Grounded Language Models: To combat hallucinations, specialized models are being trained to be strictly “grounded” in provided context, which is essential for regulated industries like finance and legal.

Agentic AI and the New RAG Paradigm

• Active vs. Passive RAG: Traditional RAG is passive, whereas Agentic RAG involves a “planner” model that actively reasons over context, decides which data sources to query, and manipulates information to find the best answer.

• Solving Complex Data Problems: Multi-agent systems can now solve “illegible” problems, such as performing root cause analysis on device log files in minutes—a task that previously took humans days.

• Specialization over Generalization: General-purpose models often “think they know better” than the provided context; specialized enterprise platforms allow for finer control over temperature and factual accuracy.

Bottom-Up Innovation and Workforce Impact

• The Power of Small Things: Real ROI often comes from empowering employees to build their own agents for unique workflows rather than only focusing on top-down cost-cutting measures.

• Job Evolution and Re-skilling: AI is expected to augment rather than replace white-collar jobs; the critical future skill will be the ability to precisely articulate thoughts through language.

• Philosophy and Communication: As “super creators,” humans must lean into fundamentals like problem framing, morality, and intuition—skills found in fields like philosophy and English literature.

Ethics and the Future of Human Connection

• The Context of Innovation: Premature regulation can stifle technology; the goal should be to innovate quickly while allowing society to develop appropriate usage principles over time.

• Heads-Up Interaction: There is hope that conversational AI will move us away from being “heads down” on small smartphone screens and toward more heads-up, social engagement with the world.

• Human Edge: Despite AI’s speed, humans maintain the “edge” in trust, connection, values, and the ability to envision a moral future.