The beginning of this story is all based on anecdotal recitation and should not be considered authoritative until someone can provide evidence of its authenticity (We assessors are like that :)
During World War II, some allied soldiers in Australia were on a firing range early in the war when one of the soldiers cocked his weapon and loaded a rifle round into the firing breach of his rifle. When he pulled the trigger, the breach exploded, giving the soldier a serious headache and a minor wound (although it could have been much more serious or even life threatening).
Following an investigation conducted by the Australian Department of Munitions, it was discovered that some samples of the lot of rifle ammunition used that day exceeded the tolerance limits for that type of ammunition and the problem was traced back to an out-of-tolerance ring gauge in the factory that had produced that lot of ammunition. This resulted in a set of QA/QC specifications which started in Australia and was eventually adopted by many Allied nations during the war.
Today, one descendent of these common specifications is known as the NATO Allied Quality Assurance Program (AQAP), a part of NATO's own supplier QA/QC requirements.
The other descendent is, anecdotally, ISO/IEC 17025. That standard grew out of the early versions of the war effort attempt at specifying laboratory QA and QC. It was taken up by national programs of most industrial nations in some form or other long before it was compiled into any international format and published by the ISO or other organisation.
The following excerpt is from an article written by Greg Gogates in 1999, following our successful completion of creating ISO/IEC 17025 as ISO/CASCO Working Group 25:
The first broad based Laboratory Accreditation Program was established in Australia as the National Association of Testing Authorities (NATA) in 1947. The second such program was the Testing Laboratory Registration Council (TELARC) of New Zealand some fifteen years later. By the late 1970s similar programs were established in the United States, United Kingdom, Denmark, France, Germany, Sweden, Canada, and several other nations.
Concern for the technical competence and quality of test and calibration laboratories and the inherent barrier to free trade when measurements are in doubt, led to the first International Laboratory Accreditation Conference (ILAC) in 1977 (sponsored by the Lab Accreditation Systems of the United States, Denmark, New Zealand, and Australia). ILAC provided input to the International Organization for Standardization (ISO) and the International Electrotechnical Commission (IEC) which represent worldwide standards development organizations (supported by almost every industrial nation).
ISO and IEC produced a joint document in the Conformity Assessment Sub Committee of the ISO (CASCO), designated as ISO/IEC Guide 25-1978, "General Requirements for the Competence of Testing Laboratories" in 1978. Subsequent versions of the standard were published in 1982 and 1990. The 1990 version of the standard specifically included calibration laboratories and, "the relevant requirements of the ISO 9000 series of standards..." in ISO/IEC Guide 25-1990, "General Requirements for the Competence of Calibration and Testing Laboratories."
It was the inclusion of considerations from ISO 9000 that lead to the changes resulting in the 1999 version of the standard, ISO/IEC 17025, from the work of our ISO/CASCO Working Group 25. Within one year, the authors of ISO 9000 had modified their standard to ISO 9000:2000 and forced all other standards within the ISO library to either remove the word "quality" from their wording or modify it to reflect ISO 9001:2000, thereby turning it over to the authors of that standard, ISO/TC 176.
CASCO elected to remove the offending wording and our working group (ISO/CASCO/WG 25) was reconstituted as ISO/CASCO/WG 10. Our committee then made a series of cosmetic modifications to the standard in the clauses dealing with management system considerations and also made two substantive changes to the clauses on competence (5.2) and quality control (5.9). The result was ISO/IEC 17025:2005.
Today the newest version of the standard is ISO/IEC 17025:2017 and there is concern that the standard is now considered more of a management system standard than a laboratory QA/QC standard. If the people who use it, who specify its use, and who accredit labs against it all appreciate that the QA/QC aspects of lab operations are the ones that actually allow a lab to produce technically valid results (still cited as the overriding rationale for the standard in its Introduction today), and that management systems only support this effort (and do not drive it), then it will continue to serve labs well.
If the focus changes to emphasise it as a management system standard, then the fears of two of the people who helped create the 2005 version of the standard will come to fruition: "Within 5 years, management system certification bodies will be certifying labs to ISO/IEC 17025." One of the speakers is still active in the international laboratory community and the other is retired. They were proved wrong in 2010, but the 2017 version of 17025 did not exist then.
It is my hope that 17025 remains a lab QA/QC standard with some management system tools added to make it easier to support the work of the people in the lab. This approach is what gives regulators comfort that 17025, and accreditation against it, remains focussed on technical competence and the validity of results produced by labs.