Dive Brief:
- IBM is enabling self-hosted deployment for its agentic software development platform Bob, letting enterprises bring the technology to data hosted on-premises, in private cloud or sovereign cloud environments, the company said Wednesday.
- IBM Bob’s deployment flexibility means enterprises managing regulated data, mission-critical infrastructure or sensitive source code will have more control as they apply AI to software delivery and modernization projects across their environments, according to IBM.
- “Organizations need AI that operates inside environments they have control over, especially when working with sensitive code and regulated data,” Neel Sundaresan, general manager of AI and automation at IBM, said in the Wednesday announcement.
Dive Insight:
IBM is betting on its ability to run AI and orchestrate agents across clouds, on-premises environments and models while also providing observability, governance and security — a growing area of concern for IT executives.
IBM Bob helps enterprises build AI applications and agents and creates a path of adoption for IBM’s agentic governance platform watsonx Orchestrate and broader AI platform, CEO Arvind Krishna told investors during the company’s Q2 2026 earnings call in July.
IBM’s AI strategy is built on “hybrid, sovereignty and trust,” Krishna said during the earnings call.
As they seek to support AI deployment, more than 80% of organizations are rethinking their cloud plans, according to an Information Services Group report. Hybrid approaches, often including a mix of public, private and sovereign cloud offerings, are gaining popularity, the report found.
Governance, particularly for AI agents, is also gaining momentum as incidents such as a rogue OpenAI agent hacking into an Australian government agency lead to security and management concerns for IT leaders. One in four AI agents operate unmonitored within enterprise environments, leading to operational risk, according to New Relic research published earlier this month.
Infrastructure giant Nvidia on Monday launched the Open Agent Safety Platform in a bid to bolster AI security and add oversight for agents. The platform and its associated technologies are being used by more than 100 companies including Microsoft and Salesforce. Meanwhile, others — including Google, AWS and OpenAI — have launched their own agentic governance platforms.
“The future of enterprise AI will depend on security, governance and sovereignty,” Sundaresan said in the Wednesday announcement.
