← Latest · Archive

SEV-3OpenAI
2 sources standard

OpenAI announced the launch of its first certification courses on 9 December 2025, marking the company's entry into formal credentialing for AI practitioners [source]. The courses target developers, product managers, and business leaders seeking structured training on OpenAI's models and APIs.

The initial certification program includes two tracks: a technical course covering API integration, prompt engineering, and fine-tuning workflows, and a business-focused course addressing use case design and deployment strategy. Both courses conclude with proctored assessments and issue digital certificates upon completion.

OpenAI stated the certifications aim to "establish industry standards for working with generative AI systems" and provide employers with a benchmark for evaluating candidate skills. The company did not disclose pricing or whether certification holders receive priority support or access to model features.

The move follows similar credentialing initiatives from cloud providers and enterprise software vendors. Microsoft, Google, and AWS have offered AI-related certifications for several years, typically tied to their respective cloud platforms. OpenAI's program focuses exclusively on its own model family, including GPT-4, GPT-4o, and the o1 series.

Industry observers note that provider-issued certifications create potential conflicts of interest, as the entity controlling model behaviour also defines competency standards. OpenAI's models have exhibited documented drift and behavioural changes across versions, raising questions about whether certifications remain valid as underlying systems evolve.

The company indicated it will update course materials quarterly to reflect API changes and new model releases. No independent accreditation body oversees the program. Enrollment opens in January 2026, with courses delivered through OpenAI's web platform.

Why this is an AI incident

Launch-archive bulk classification (10 May 2026). Source signal originates from a real AI provider, regulator, or model-comparison probe; the harm or behavioural change described would not have occurred without the AI system being deployed in the role described. Editor reviewing the archive may amend the rationale per-wire.

Counterfactual "but-for" test per the Editor's Guide.

Codes M1, F10
Providers OpenAI