"AI agents have rapidly moved from experimental prototypes to a mainstream strategic priority inside enterprises. In his ODSC AI West 2025 keynote, João Moura offered a clear window into this shift - grounded not in speculation but in usage data, real deployments, and the recurring challenges he sees across hundreds of organizations. The headline number alone signals the scale of the transformation. João Moura's platform now runs 450 million agent executions per month, a dramatic leap from roughly 60 million a few quarters back."
"Moura grounded the audience in a simple definition: agents introduce agency. Instead of responding to single-step prompts, they reason about goals, select tools, navigate uncertainty, and take action across systems. This combination unlocks forms of automation that businesses have long wanted but couldn't achieve with rules-based workflows. Agents can generate content, handle ambiguity, recover from API failures, and complete multi-step processes end-to-end. They can schedule meetings, process reimbursements, evaluate customer risk, or prepare documents - all while adapting in real time when conditions change."
AI agents have transitioned from experimental prototypes to mainstream enterprise priorities, driven by large-scale real deployments and usage growth. Platform executions rose to 450 million agent runs per month from roughly 60 million a few quarters earlier, with Q3 activations surpassing 1.1 billion. Agents introduce agency: they reason about goals, select tools, navigate uncertainty, and act across systems rather than respond to single-step prompts. That capability enables end-to-end automation previously blocked by messy data, inconsistent logic, or unreliable external services. Agents can generate content, handle ambiguity, recover from API failures, and complete multi-step processes while adapting in real time.
Read at Medium
Unable to calculate read time
Collection
[
|
...
]