Agent Gravity : Who's Running Your Agents

2026-05-26 · Source: Tomasz Tunguz · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure, Data Science & Analytics · Depth: Intermediate, quick

Summary

Agent Gravity" is introduced as a critical force in the "Decade of Agents," mirroring "Data Gravity's" influence in the "Decade of Data." This concept describes how AI agents, due to their substantial compute requirements, attract and consolidate workloads on specific platforms. Major platforms will actively compete to host these agents and their associated data, increasing their "agent gravity." A recent example involves a new Databricks feature on Microsoft's platform, which enables Power BI customers to manage data and build AI agents within Databricks, potentially diverting data management and AI agent development away from Microsoft's competing Fabric offering. This dynamic allows agents, or their creators, to decide where to process data and run workloads, leading to the migration of profitable agent and data warehouse operations to alternative platforms.

Key takeaway

For AI Architects evaluating platform strategies, "Agent Gravity" highlights a critical shift: your choice of agent development and deployment platforms directly influences data and workload residency. Teams running AI agents should actively monitor where their agents are processing data and consider the long-term implications for platform lock-in or migration. Proactively assess agent portability and data egress costs to maintain control over your valuable data assets and avoid unintended platform shifts.

Key insights

Agent Gravity, driven by AI agents' compute demands, dictates where data and profitable workloads reside, shifting platform power.

Principles

• Agent compute demands create platform competition.
• Greater agents/data increase platform "gravity."
• Agents can siphon knowledge and migrate workloads.

In practice

• Build AI agents in Databricks for Power BI data.
• Migrate data warehouse workloads via agents.
• Publish data to diverse BI systems using agents.

Topics

• AI Agents
• Agent Gravity
• Platform Competition
• Data Management
• Databricks
• Workload Migration

Best for: CTO, VP of Engineering/Data, AI Product Manager, Director of AI/ML, AI Architect, Consultant

Related on AIssential

• Perplexity announces hybrid AI system that decides what runs locally or in the cloud — Perplexity's hybrid AI orchestrator intelligently routes tasks between local and cloud models for optimal privacy and efficiency.
• Codex Isn’t Just for Coders Anymore. OpenAI Is Coming for the Entire Office. — The expanding non-developer use of Codex is driving OpenAI's release of new enterprise features.
• Inside Google’s System for Coordinated A/B Testing Across Its Global Service Fleet — Google's system standardizes large-scale A/B testing across distributed services, ensuring statistical rigor and minimizing interference.
• Article: Two Misconfigurations That Caused Spark OOM Failures on Kubernetes — Cloud migrations can introduce silent infrastructure misconfigurations that cause Spark OOM failures under production load.
• You Can Finally Build Your Own LLM. Here’s Why You Probably Shouldn’t. — The build-versus-buy decision for LLMs is primarily an arithmetic problem, often favoring API rental over self-hosting.
• Introducing workspaces for Lambda Cloud — Lambda Cloud "workspaces" provide logical containers for GPU resources, enhancing access control and environment separation.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Tomasz Tunguz.