Literature/202504281128 certification for sdls

• Source: [[@canty2025Science acceleration and accessibility with self-driving labs]]
• Tags: #self-driving-lab #sdl #quality-control

One of the challenges of SDL (self-driving labs) is that the quality of the data they produce may be contested. The only way to move forward is to create a standardized set of quality control procedures that can be run through different labs, and which outcomes are publicly shared and periodically updated.

This quality control procedure is fundamental for centralized facilities (202504281112 SDL approach to distributed discovery) and it may be the result of establishing appropriate consortia with enough visibility and mandate to establish those fundamental rules.

In a distributed system, the only way to trust each other is through transparency based on rigorous analysis. It is also important that engaging with different stakeholders can be viewed as a strength, especially if it prevents an unreasonable increase the skill-floor required to engage with SDL (self-driving labs) type of work.

Creating a certification body for SDL's at some point will become a necessity, and there'll be a power struggle in order to be the dominant player in the field.