ISO 18587 vs 17100 vs 9001: What’s the Difference in an AI Translation Workflow?

ISO 17100 governs human translation workflows, ISO 18587 defines how machine translation is post-edited by qualified linguists, and ISO 9001 covers overall quality management systems. If your workflow includes AI translation, ISO 18587 is essential for ensuring structured, human-reviewed output. In regulated industries, using the right combination of these standards reduces risk and creates a defensible localization process.

You’ve decided AI translation might be part of your localization strategy. Now procurement is asking about ISO certifications. You see references to ISO 17100. Then ISO 18587. Sometimes ISO 9001. They sound similar. They are not. If you’re managing localization in healthcare, manufacturing, or regulated industries, understanding ISO 18587 vs ISO 17100 vs ISO 9001 helps you lower operational risk.

In this article, you’ll learn:

• What ISO 17100 governs
• What ISO 18587 adds for AI workflows
• Where ISO 9001 fits in
• Which certification matters based on your risk profile

Interpro holds multiple ISO certifications, including ISO 18587 for Machine Translation Post-Editing (MTPE). We can help ensure that your human, AI-driven, or hybrid localization workflows are built on structured processes, qualified linguists, and human-in-the-loop quality assurance.

ISO 18587 vs ISO 17100 vs 9001

• ISO 17100 governs traditional human translation services.
• ISO 18587 defines requirements for the human post-editing of machine-translated content (MTPE).
• ISO 9001 governs the overall quality management system.

If AI is part of your workflow, ISO 18587 provides a structured framework for executing and validating machine translation post-editing workflows.

Side-by-side comparison of ISO 17100, ISO 18587, and ISO 9001 showing how human translation, AI post-editing, and quality systems differ

Start with the Foundation: What ISO 17100 Covers

ISO 17100 is the gold standard for human translation services.

It defines:

• Translator qualifications
• Defined translation process stages
• Project management responsibilities
• Quality control procedures

ISO 17100 assumes translation begins with a human linguist, not machine output. If your workflow starts with a human translator creating the first draft, ISO 17100 applies cleanly. It ensures documented processes, qualified professionals, and revision requirements.

But ISO 17100 does not define how to govern machine-generated output. That’s where the gap begins.

What ISO 18587 Provides for AI Workflows

ISO 18587 was created specifically to govern Machine Translation Post-Editing (MTPE). Machine-generated translations must be revised by qualified human linguists before final delivery.

While ISO 17100 applies when translation begins with a human linguist, ISO 18587 governs workflows where the first draft is produced by a machine translation system. The standard requires full human post-editing to ensure the final translation is accurate, usable, and appropriate for its intended audience and purpose.

Post-editors systematically revise machine output to:

• Correct unclear or incorrect meaning
• Ensure no information is added or omitted
• Align terminology with approved glossaries
• Fix grammar, syntax, and semantic errors
• Apply appropriate tone and stylistic guidance
• Maintain formatting, tags, and structural integrity

A final validation step confirms the translation is suitable for real-world use. Post-editors may also provide feedback on machine translation performance to support ongoing workflow improvement.

While MTPE can deliver strong consistency and efficiency, the final output may, depending on content and machine quality, lack some stylistic nuance achieved through fully human translation. ISO 18587 introduces defined responsibilities and validation expectations that bring structure and accountability to AI-assisted localization workflows.

If you are implementing AI translation or MTPE, ISO 18587 becomes the formal framework that governs that process.

Without it, you are relying on an internal interpretation of what “revision” means.

Where ISO 9001 Fits in the Stack

ISO 9001 does not govern translation itself. It governs your quality management system (QMS).

That includes:

• Process documentation
• Corrective action procedures
• Continual improvement
• Risk management systems
• Audit controls

If ISO 17100 and ISO 18587 define how a translation is executed, then ISO 9001 defines how your organization manages quality across services.

If you operate in a highly regulated industry like healthcare, life sciences, or manufacturing, a layered and controlled approach matters. It signals maturity, accountability, and process discipline.

Side-by-Side Comparison: ISO 18587 vs ISO 17100

This comparison clarifies the structural differences. The distinction lies in how the first draft is created and how that draft must be validated.

Why This Difference Matters in Regulated Industries

If AI touches:

• Clinical documentation
• Patient-facing materials
• Financial disclosures
• Safety instructions
• Technical manuals

Then controlling the localization process must extend to cover the challenges of AI machine output. For example, a life sciences organization implementing AI translation for regulatory submissions cannot rely solely on ISO 17100. That standard assumes human-originated content.

ISO 18587 formalizes:

• How post-editors are qualified
• How errors are categorized and corrected
• How machine output is elevated to acceptable quality

n regulated sectors, this is a governance requirement, not just a theoretical distinction.  Without ISO 18587, there is no formalized post-editing framework applied to AI output. That is a governance gap.

Common Misunderstandings About ISO Translation Standards

Myth 1: ISO 17100 Covers AI Translation

Reality: ISO 17100 does not define the post-editing of machine translation output. 

The standard was created before the rapid expansion of AI-driven translation tools. As a result, it does not address how machine-generated translations should be reviewed, revised, or validated.

ISO 18587 clearly defines the international standard for Machine Translation Post-Editing (MTPE). It outlines how professional linguists should revise machine translation output to ensure that the final content is accurate, fluent, and faithful to the source text. The standard also establishes requirements for post-editor qualifications, quality control, and workflow structure.

In short:

• ISO 17100 governs traditional human translation workflows.
• ISO 18587 governs the post-editing of machine translation output.

Myth 2: MTPE Is Just “Light Revision”

Reality: ISO 18587 requires full post-editing according to predefined linguistic and quality standards.

One of the most common misunderstandings about Machine Translation Post-Editing (MTPE) is that it simply means someone is quickly skimming an AI translation to catch obvious errors. Or that you can opt for a “light MTPE” option. 

In reality, professional MTPE is a structured linguistic process, not a casual review.

Under the ISO 18587 standard, post-editors are trained linguists who are responsible for ensuring the machine-generated translation meets a certain quality criterion.  The goal is not just to make the text “good enough.” The goal is to ensure the final translation is complete, accurate, grammatically correct, and appropriate for its intended audience and purpose.

That means post-editors are expected to correct grammar, syntax, and readability issues created by machine output while also verifying that terminology aligns with approved glossaries and industry standards. In addition, post-editors ensure that the meaning of the source text is preserved without distortion, remove any literal translations that sound unnatural in the target language, and adjust tone and phrasing so the content reads smoothly and naturally for the intended audience.

Myth 3: ISO Certification Guarantees Zero Errors

Reality: ISO certification guarantees a defined process, documentation, and accountability. It does not eliminate human or system risk.

A common misconception about ISO certification is that it guarantees perfect outcomes. In reality, ISO standards are not designed to yield zero errors. Instead, they establish structured processes and quality controls that reduce risk and make translation workflows more reliable, transparent, and repeatable.

When a translation provider is ISO certified, it means their workflow follows documented standards that govern how work is performed, reviewed, and delivered. These standards define requirements such as linguistic qualifications, revision procedures, and quality management practices.

However, translation (whether human or AI-assisted) is still a complex linguistic task. Language is contextual, nuanced, and influenced by industry-specific terminology, cultural factors, and audience expectations. Because of this, no certification can eliminate every possible risk or error.

Which Certification Should You Require?

Your answer depends on your workflow and risk profile.

If you use only human translation: ISO 17100 and ISO 9001 are sufficient.

If you use AI + human post-editing (MTPE): ISO 18587 is critical.

This layered approach ensures how human, AI, and organizational quality systems are governed. 

For enterprise buyers navigating AI adoption pressure, ISO 18587 provides defensibility. And defensibility is the new currency in AI-driven localization.

Reliable AI translation requires structured workflows that define how machine-generated content is reviewed, revised, and validated. ISO standards provide the governance framework that makes this possible.

Need Help Building Your AI Localization Workflow?

If you’re evaluating AI translation for localization workflows, the first step is understanding your content risk levels and governance requirements.

Interpro helps organizations design structured AI translation systems that balance efficiency with compliance, accuracy, and quality control.

Schedule a consultation to assess your workflows and explore how ISO-aligned processes can support responsible AI localization.

Identify where ISO 17100, ISO 18587, and ISO 9001 fit into your risk model.

FAQs

Does ISO 17100 include machine translation?

No. ISO 17100 governs human translation services and does not define MTPE requirements.

Is ISO 18587 required for MTPE?

If you are formally implementing machine translation post-editing in enterprise or regulated environments, ISO 18587 is the recognized standard governing that process.

Can a vendor offer MTPE without ISO 18587?

Yes, but without ISO 18587 certification, there is no externally-validated framework defining how post-editing is executed.

Does ISO 9001 replace translation standards?

No. ISO 9001 governs quality management systems. It does not replace ISO 17100 or ISO 18587.

Should healthcare companies require ISO 18587?

If AI touches regulated, patient-facing, or compliance-driven content, ISO 18587 provides structured governance for machine-originated translations.