Fair trial in the torus of digital managerialism of the criminal system

The discussion of civil liability arising from algorithmic losses - which is mostly investigated as civil liability of artificial intelligence or liability arising from the use of artificial intelligence is nascent /aborning in the legal literature. Thinking algorithms require a special civil liability system in case of losses due to the unique characteristics of self-learning, randomness, unpredictability and autonomy. The general theory of Opposability can be a guide in such new cases, too. However, having a transparent and pragmatic criterion and standard to compare the harmful algorithm with similar cases to determine its normal operation is associated with challenges. Determining the scope of the concept of reasonable algorithms, in terms of their unknown nature and inherent complexities, needs to be investigated. By examining the existing obstacles, this research pursues the re-examination of reasonable algorithms with a mixture criterion: A standard that creates incentives for technological and innovative spaces, compensates for the damage, improves the safety level of algorithms and realises the appropriate flexibility to face new generations of technology. Rereading the Opposability theory (causality customary) in the field of algorithms with conventional and reasonable efficiency, as well as adapting this general theory in dealing with the losses arising from the use of thinking algorithms, is the main result of ​​the authors' efforts in this article.