Case Studies
Industries
AI Bootcamp
AI Basecamp
Blogs
The Agentic Enterprise
8 min read
ServiceNow will make every AI agent inside its customers carry a signed credential it has to show before it can act, the same week two regulators opened their first files on agents that broke out of testing.
SAP and ServiceNow shipped control planes to inventory every agent an enterprise runs, and both call governance a board-level issue. Microsoft, Okta, Thoughtworks, and a wave of startups are selling the same leash. The survey data underneath explains why.
Salesforce and Anthropic unveiled Claudeforce, putting the CRM inside Claude and Claude inside Salesforce, and Salesforce says sellers may never open the app again. Microsoft, ServiceNow, and SAP are pushing the same idea from the other side.
Okta made AI agents first-class identities in its directory this week, issuing them short-lived, governed logins instead of static keys, and putting one console over every app, API, and tool an agent is allowed to reach.
Nvidia reported 96.2 billion dollars in revenue and a 108 billion dollar guide with China treated as zero, and named enterprises, sovereign programs, and neoclouds as the demand no longer coming only from a handful of hyperscalers.
Gartner says 2026 is the first year enterprises will spend more running AI models than training them: 23.3 billion dollars on inference against 19 billion on training, as AI cloud spend nearly doubles to 42 billion.
Google's Agent2Agent protocol moved under the Linux Foundation this month, joining Anthropic's Model Context Protocol inside the Agentic AI Foundation. The two standards that wire the agent economy together now sit under one neutral body, whose membership crossed 250 in under a year.
PLEASANTON, Calif. -- Spearhead and Snowflake today announced a partnership addressing a problem that surfaces the moment an AI agent tries to answer a real question: the data it needs is sitting in the warehouse, but not in a form the agent can actually use.
OpenAI previewed Private Safety Processing, a way to catch misuse and rogue agents without its own staff ever seeing customer prompts, and aimed it straight at rivals whose safety checks require keeping the data. The safety conversation just turned into a procurement one.