Compliance
Compliance (regulatory compliance) means conforming to comply with relevant laws and regulations. Together with KYC Spider in Zug Eurospider has in-depth compliance know-how.
Machine learning is widely used in computer science and other fields such as compliance. Developing successful machine learning applications, however, requires a substantial amount of "black art" that is difficult to find in textbooks. Here come the 12 key lessons that machine learning researches have learned.
It’s a familiar scenario. You need to print something in a hurry, but the printer isn’t working – and neither is the ticket machine. And the train is has come to a halt somewhere along the tracks, probably due to a software bug. Unfortunately, we still haven’t found a way to get rid of these critters once and for all. In 1977, Caspers Jones estimated that for every 1,000 lines of code, one can expect a specific number of bugs that cause incorrect results or unexpected behaviors in computer programs. So from a compliance point of view, bugs are always to be expected. What does this mean?
White and black lists, also called positive and negative lists, are used for binary categorizations. Examples are spam filters against unwanted e-mails, but also proscription lists in the Roman Empire with the names of unpopular political opponents to be murdered. More recent are the lists published by the American President George Bush after the terrorist attacks on September 11, 2001. White and black lists can be combined in different ways: All positives but without negatives and all negatives but without the positives. In compliance, whitelisting is often used when periodically checking a customer base.
On December 13, 2019, an interdepartmental work group presented its report Challenges of Artificial Intelligence commissioned by the Federal Council. Section 1.6.16 addresses the use of artificial-intelligence in the judicial system. The matter of compliance is also relevant here. The European Ethical Charter on the Use of Artificial Intelligence in Judicial Systems and their environment distinguishes between four types of use: “uses to be encouraged”, “possible uses, requiring considerable methodological precautions”, “uses to be considered following additional scientific studies”, and “uses to be considered with the most extreme reservations”. What does this mean specifically in terms of compliance?
Assistance systems exist in a variety of forms. The simplest passive systems are limited to recommendations and warnings, which people can either heed or ignore. Digital assistants that carry out conversations are more complex. The easiest ones to develop are digital assistants that control the conversation – yet these assistants are anything but trivial. The assistant asks questions and the person answers them. This makes it possible to formulate questions that based on previous answers. More sophisticated assistants bring users out of an “anomalous state of knowledge” (ASK).
Complete Revision of the Federal Data Protection Act
Complete Revision of the Federal Data Protection Act: „As of 15th September 2017, draft and report for a completely revised Federal Data Protection Act is public. In a first step parliament and the people agreed to adaptations in order to be compliant with EU law. The second part of the revision is debated by the parliament since September 2019. Data Protection is to be increased by giving people more control over their private data as well as reinforcing transparency regarding the handling of confidential data.”
Links: datenrecht.ch
