Latest deployments show AI agents trained through machine teaching are delivering measurable impact and moving towards autonomy

SAN FRANCISCO — Dec. 16, 2025 — AMESA, the enterprise AI agentic orchestration and proving ground platform, announced today that it has delivered more than $100 million in combined realized and projected annual value across Fortune 500 enterprises. As enterprises search for real ROI from AI, the milestone reflects accelerating adoption of Physical AI, where teams of AI agents safely control real operations, rather than simply generating insights. 

While many companies are still experimenting with single-model AI tools, AMESA enables enterprises to deploy teams of autonomous agents that learn expert reasoning, coordinate actions and improve through structured practice inside safe digital environments. These agents are trained using machine teaching, the methodology pioneered by AMESA founder and CEO Kence Anderson, which allows human experts to directly impart skills, strategies, and decision rules to AI systems.

“Enterprises want AI that delivers real outcomes, not more dashboards,” Anderson said. “For years, companies have invested in systems that analyze operations but do not actually run them. AMESA solves that gap by providing enterprises AI agents that operate physical systems with the judgment of the human experts who train them. This is why we are seeing rapid and measurable value.”

Powered by AMESA’s platform suite, including Agent Cloud, Agent Orchestration Studio and Assist Agents, recent multi-agent deployments are delivering significant operational and financial gains:

• Oil refinery processing: AMESA modeled a refinery’s crude oil blending network, one of the industry’s most economically sensitive operations. Blending requires meeting strict product specifications while navigating dozens of operational constraints. Using this digital environment, AMESA trained a coordinated team of agents to reason about blend quality, safety limits and margin tradeoffs. The deployment increased daily blending profit by 16%, totaling more than $21 million per year at a single refinery, with potential for multipliers across additional sites.
• Industrial technology provider: Working inside a digital twin of a nitrogen manufacturing process, AMESA trained multiple agent formations to optimize energy use and production economics. Within 24 hours, all formations outperformed the system’s industry-leading Model Predictive Controller (MPC). The top formation is projected to generate $1.2 million annually.
• Global glass manufacturer: One of the world’s largest glass-container producers partnered with AMESA to automate a critical production step that previously relied on operator intuition. Using machine teaching, senior operators transferred decades of tacit knowledge into a coordinated team of eight agents. With nearly 25% of the manufacturing workforce now over 55, this deployment marks a major step toward preserving expert reasoning, ensuring consistent operations and advancing toward autonomous process control.

An executive from the glass manufacturer shared, “The AMESA platform allowed us to combine our data and operator expertise to develop an intelligent autonomous agent capable of making decisions at an expert level.”

The Path to Wide-Scale Enterprise Adoption of AI Autonomy

The company’s $100 million milestone signals a broader enterprise shift from experimental AI tools to AI autonomy, where AI systems do real work inside real operations.  

AMESA’s approach ensures agents practice in digital twins and gain clarity, accountability and skill like human experts. This combination of orchestration and practice enables enterprises to scale AI safely and sustainably.

“Expertise has always been built through practice, never theory, and our agents follow this principle,” Anderson said. “Autonomy only becomes possible when AI is designed the way enterprises already design human expertise.  That’s the path to wide-scale enterprise adoption and lasting value.”

AMESA is continuing to expand deployments across additional Fortune 500 operations in the energy, manufacturing and consumer sectors. To learn more and partner with AMESA, please visit amesa.com. 

About AMESA

AMESA is the platform for AI agentic orchestration and practice, providing the core infrastructure for deploying autonomous AI teams that operate real physical systems. Founded by Kence Anderson, creator of the machine teaching methodology and author of Designing Autonomous AI (O’Reilly), AMESA enables Fortune 500 companies to achieve measurable operational autonomy in manufacturing, energy, logistics, and industrial processing. The company is headquartered in San Francisco with teams across North America, South America, and Europe. To learn more, please visit www.amesa.com

SOURCE: http://www.intelligence360.io
Copyright (c) 2025 SI360 Inc. All rights reserved.