> ## Content Index
> Fetch the complete content index at: https://www.ai-360.online/llms.txt
> Use this file to discover other available public pages before exploring further.

# IBM gives enterprise AI agents distinct identities across rival platforms
- URL: https://www.ai-360.online/ibm-gives-enterprise-ai-agents-distinct-identities-across-rival-platforms/
- Published: 2026-10-06T05:50:00.000Z
- Updated: 2026-10-31T21:45:04.000Z
- Description: IBM watsonx Orchestrate now discovers agents on Microsoft Foundry and Google Gemini Enterprise, and previews Agent Identity to give each agent its own identity.
- Author: AI-360
- Tags: 2. Fraud & Identity, 1. Enterprise AI, News, identity, Agentic AI, #industry-fraud-identity

[IBM](https://www.ibm.com/?ref=ai-360.online) has expanded watsonx Orchestrate so its AI Gateway can discover and import agents built on [Microsoft](https://www.ai-360.online/microsoft/) Foundry and Google Gemini Enterprise Agent Platform, and has previewed Agent Identity, which gives each agent its own verifiable identity, [in a product update published on 2 October 2026](https://www.ibm.com/new/announcements/new-in-ibm-watsonx-orchestrate-expanded-third-party-agent-management-and-agent-identities?ref=ai-360.online).

The gateway already connected to Amazon Agentcore. [IBM](https://www.ai-360.online/ibm/) said teams can scan each connected platform, discover the agents running on it and govern them in the same control plane as native watsonx Orchestrate agents. The cross-platform discovery preview was released on 15 September 2026.

Agent Identity gives each agent an identity separate from the person who built it and the person using it, managed through the identity provider a security team already uses.

IBM also made six prebuilt LLM-as-a-Judge evaluators generally available on 30 September for live agent traffic: toxicity, helpfulness, hallucination, conciseness, context relevance and answer relevance. Admins can switch evaluators on or off across a tenant and set sampling rates from a 3% default up to 20% per evaluator, which IBM described as a direct cost lever because each evaluation is a model call.

IBM did not give a general availability date for Agent Identity.

---