If you use or produce artificial intelligence solutions in your organization, you need to establish a AI regulation advisory group as soon as possible. It's not just good to have again. The European Commission published the first legal framework proposal in AI, which showed that anyone who spread AI must comply with national and European specific regulations - or face a large fine. Every EU member country already has, or is working, a national AI strategy that will combine this framework and will change it into a national law.
Also Read: IT support
Company in Pune
AI regulation problems are not only relevant at the European level. On January 13, 2020 The Design of Design A.S. M.S. Publishes for artificial intelligence regulations (AI). It is clear that the production and use of AI will be regulated throughout the world in the next few years.
So why does Ai need regulation?
Artificial intelligence is still a discipline that arises in the IT industry and wider software, offering great potential in terms of impact. This is the promised impact on the community which makes it very valuable, but also very dangerous.
Also read: IT support Company in Bangalore
Building and mobilizing a safe AI is the most important, so the rules cannot be avoided (similar to how we implement it in the industry regulated such as nuclear, finance, or energy). One year ago I debated in Linkedin's post that AI's explanation (XAI - explained AI) very connected with AI regulations because AI regulations came from the fact that AI technology was not deterministic (meaning you could not predict the results of the new input data set), and that the system The deepest neural network is a "black box" that does not predict results based on input data, or understand how the results are achieved.
Also read: IT support Company in Noida
Now, imagine that the AI system has the opportunity to produce unwanted behavior because poor processes or supervision is responsible for determining your credit assessment, determining the municipal tax, resulting in medical imaging results or marking you for law enforcement as a potential future crime, " Minority report "-style. Beyond the impact that AI has, and more and more, in the community and its citizens, we can also reflect on the "weaponry" of AI and its use in information on online disinformation, unmanned military vehicles, cyber warfare, and so on.
Artificial intelligence, such as other software solutions that have an impact, requires supervision of development to production and its surroundings: from the collection and management of datasets, building machine learning models, deployment, use, human failafe approach, output accuracy and maintenance, and user education. Every aspect of AI needs to adhere to a series of important ethical guidelines that are generally transferred to rules and legislation, so the AI does not end or used for the wrong thing.
Also read: IT support Company in Chennai
In short, the UE AI rules in short
Regulations proposed by the European Commission have extensive scope, which means providers, importer, distributors and users are all affected. AI systems will be classified by risk: unacceptable, high risk, limited risk, and minimal (not regulated). This strictly prohibits the use of AI for all types of social assessment by any public authority, and the use of biometric identification systems in public spaces targeting the general public. High-risk AI systems include software to recruit, determine access to social benefits, controlling migration, and helping judicial interpretation. Although it is still unclear and incomplete, a high-risk AI list is set to be expanded further through the EU database for high-risk AI practices.
Also read: IT support Company in Mumbai
All documents focus on the fundamental values of the European Union and how the AI system can disrupt or limit the rights of its citizens, from exploiting certain groups of vulnerabilities to be treated unfairly and unprofitable.
Also read: IT support Company in Delhi
No comments:
Post a Comment