Summary
The panel argued that agentic AI delivers value when models receive business, market, user, policy, and governance context—not just more raw intelligence. Bianca described Ailey’s semantic-layer approach as producing 700 times more accurate recommendations and up to 70% cost efficiency, while Stefan emphasized APIs, access management, measurement, latency, and ROI as enterprise deployment foundations.
Key Topics
5 key topics from Bianca Anghelina, Stefan Ostwald and Jeff Wilser at HumanX Amsterdam 2026. Thicker branches were mentioned more often in the talk.
The Missing Ingredient in Agentic AI: Context
Click a topic to open it
Key Takeaways
- Prioritize rich business context over simply chasing bigger models.
- Structure internal, external, and user context before deploying agents.
- Capture real-time decision feedback where users already work.
- Measure ROI and latency, not token cost alone.
- Use focused optimization goals to prevent runaway reasoning loops.
“I would argue in the AI space now, context is king in many in many ways.”
Share
Up Next
DJA
WatchNext in agenda
The Autonomous Finance Office
Daniel Kim, John Glasgow, Anna Tutova
autonomous financeai governance
SAN
WatchAlso on agentic ai
An Agentic Backbone
Steve Lucas, Arun Chandra, Nidhi Aggarwal, Tomas Statius
agentic aiai governance
PAD
WatchAlso on agentic ai
Context is King: How to Build AI Systems, Not Just Features
Prukalpa Sankar, Ariel Shulman, Dana Lawson, Emil Protalinski
agentic aicontext engineering