IBM embeds GPT-5.6 and OpenAI Codex into its consulting platform in a new enterprise AI partnership

IBM and OpenAI announced a strategic partnership on 13 August 2026 to embed OpenAI frontier models - including GPT-5.6, Codex, and ChatGPT Work - into IBM Consulting Advantage, the company's enterprise AI delivery platform^[1]. Financial terms were not disclosed^[1].

What goes into IBM Consulting Advantage

IBM will incorporate GPT-5.6, Codex, and ChatGPT Work into IBM Consulting Advantage, combining them with IBM's existing consulting assets, industry solutions, and security capabilities^[1]. The platform is designed to analyze operating procedures, identify inefficiencies, and support AI-based automation across enterprise workflows^[1].

IBM will also establish a dedicated OpenAI Practice. Thousands of IBM consultants and engineers are expected to gain advanced certifications through the OpenAI Partner Network, with specialized forward-deployed teams assigned to customers in complex and highly regulated environments^[1]. IBM will join OpenAI's Elite partner tier as part of the agreement^[1].

The joint go-to-market effort targets four industries first:

• Financial services
• Government
• Telecommunications
• Retail

Within those sectors, the focus spans finance, procurement, customer operations, and HR^[1].

Application modernization and software engineering

A second workstream pairs Codex and ChatGPT Work with IBM's technology expertise to simplify engineering processes and accelerate development of digital products and services^[1]. The companies frame this as a path to modernizing legacy applications - a persistent challenge for large regulated organizations that still run significant workloads on older infrastructure^[1].

Cybersecurity via OpenAI Daybreak

The cybersecurity component builds on IBM's participation in the OpenAI Daybreak Cyber Partner Program, which the two companies launched in June 2026^[1]. Under the expanded agreement, OpenAI's AI capabilities will be combined with IBM Autonomous Security, a multi-agent service that coordinates security decision-making, response, and intelligence^[1].

The collaboration targets three specific risk categories:

• Application-layer vulnerabilities
• AI governance gaps
• Operational issues that slow enterprise adoption^[1]

Context

The deal comes less than a year after IBM announced a similar alliance with Anthropic, and fits IBM's broader model-agnostic strategy of pairing its own Granite models with third-party frontier AI through its watsonx platform and consulting business. IBM lowered its 2026 revenue forecast last month following weaker-than-expected quarterly results, and the OpenAI partnership represents one of the more visible moves to accelerate growth in its AI services segment.

The near-term test is whether the dedicated OpenAI Practice can convert certifications into billable engagements - and whether IBM Autonomous Security, combined with Daybreak, gains traction in the regulated sectors the partnership explicitly targets.