Data Privacy Risk In Artificial Intelligence

Posted By :Arun Singh |26th October 2020

                                                 
                                                      Image downloaded from: https://www.nist.gov/sites/default/files/images/2019/07/02/composite_hero.jpg
 

A basic privacy statement has the power to conceal, or infer about your personal information, to reduce the limitations that others may have in our conduct. Privacy is traditionally respected as a requirement for the exercise of human rights such as freedom of expression, freedom of assembly, and freedom of choice.

In the case of data, confidentiality depends on our ability to control how our data is stored, modified, and exchanged between different groups.
With the advent of high-level online-based mining strategies in recent decades, privacy has become a relevant social problem. Social actors who regularly use these methods, such as government agencies and companies, are now in a position to identify, prophesy, and directly touch people's lives without their consent. And with the advent of artificial intelligence systems that are becoming more and more complex, these private concerns have only intensified.


What is required in the corresponding privacy policy
 

To determine if anonymity does not occur in the context of your business, the next step should be to obtain permission for data topics. This can be tricky, especially in situations where the basic data is collected anonymously.

Many companies rely on privacy policies as a means of obtaining permission to collect and process personal information. For this to work, privacy policy must clearly define and, in particular, how data should be used. It is common to say that data can be used to train algorithms that are generally inadequate. If your data scientists discover new uses of the data you have collected, you should return to the data studies and make them comply with the updated privacy policy. The FTC considers the company's non-compliance with its privacy policy to be an unreasonable trading practice based on possible investigations and penalties. This type of non-compliance was the basis for a $ 5 billion fine imposed on Facebook last year.


Consider how and when data can be identified
 

Privacy laws apply to the control of personally identifiable information. When a person's identity is not known, most privacy issues disappear. That means data data is often based on the ability to identify the person you are associated with, or at least to be able to combine different data sets that are almost identical to the same person.

Computer scientists may see a so-called hash method of one-way hash as the method of anonymity used to train machine learning algorithms. Hash functions work by converting data into numbers in such a way that the original data is not only available to this number. For example, if a data record has the word "John Smith" associated with it, the hash function can convert the word "John Smith" into numbers which are difficult or impossible numbers to find that person's name. This method of anonymity is widely used, but it is not stupid. European data protection authorities have issued a detailed guide on how hashes can be used and cannot be used to make information anonymous.Consider how and when data can be identified
 

                                                 
                                      Image downloaded from: https://venturebeat.com/wp-content/uploads/2019/11/GettyImages-1079012962-e1574180775599.jpg?w=1200&strip=all
 

AI Privacy Issues
 

AI has challenged the ability to look and feel great. The ability to train a comprehensive learning program (DL) in large amounts of data has increased the speed of analysis and results, but the need for more information increases the risk of lack of privacy. To provide procedures for dealing with that challenge in a timely manner, software can also help.

The challenge for Congress is to pass a privacy policy that protects people from any adverse effects on the use of personal information in AI, but without unnecessarily limiting AI development or holding privacy laws in complex social and political contexts. AI interviews in the context of confidential talk often bring limitations and failures of AI programs, such as speculative police can affect minor differences or failed Amazon tests with a hiring algorithm that replicates the company's unequal existing male employees. Both of these raise important issues, but privacy law is complex enough even without packing into all the social and political issues that may arise from the use of information. To assess the impact of AI on privacy, it is necessary to distinguish between data problems that exist across AI, such as situations of negative and negative attributes or patterns, and those that are directly related to the use of personal information.


Tips for board members dealing with AI
 

  • Encourage managers to separate AI from other technological risk analysis in order to disclose confidential information the technology creates and any potential data risks.
     
  • Make sure safety rules are followed by vendors long after service contracts have been signed. Encourage managers to keep standard schedules to ensure that technology partners keep their promises to protect their information.
     
  • Push management complies with a very strict set of privacy laws, whether the company is currently not in the EU or in other markets with far-reaching needs. That way, if the company expands to those areas, it will not be a major burden to re-implement security policies.
     
  • Monitor technical contractors to ensure compliance with safety regulations. If the AI-based tool AI has to remove external data, ask for confirmation that such removal is possible. 


How AI Can compromise Privacy


What made AI attractive to use in early data collection were three factors: speed, scale, and automation. The speed at which AI performs statistics is already faster than what human analysts can do, and it can increase illegally by adding more tools.

AI is also naturally capable of using large data sets for analysis, and is apparently the only way to process large data in a limited time. Finally, AI can perform set tasks without monitoring, which greatly improves the efficiency of analysis. For example Voice recognition and facial recognition are two ways to indicate that AI is increasingly becoming more efficient at work. These methods have the potential to anonymously threaten the name of the public sector. For example, law enforcement agencies may use facial recognition and voice recognition to locate people without a valid reason or reasonable suspicion, thus avoiding legal procedures they should have followed.
 

Conclusion


Digital technology such as AI has made a huge difference in many areas of our lives. The vast amount of information we can collect and analyze using these tools allows us to deal with social problems that were previously unresolved.

Unfortunately, this technology can be used against us by a variety of social actors, from individuals, to companies, to government agencies. Our loss of privacy is just one example of how AI-like technology works in our environment. However, if we can better understand these technologies, and their impact on our daily lives, we will find ways to protect ourselves from exploitation by those who use them for malicious purposes.
 


About Author

Arun Singh

Arun is a MEAN stack developer. He has a fastest and efficient way of problem solving techniques. He is very good in JavaScript and also have a little bit knowledge of Java and Python.

Request For Proposal

Sending message..

Ready to innovate ? Let's get in touch


Notice: Undefined index: HTTP_REFERER in /var/html/www/AI/wp-content/themes/oxides-child/functions.php on line 272

Notice: Undefined index: HTTP_REFERER in /var/html/www/AI/wp-content/themes/oxides-child/functions.php on line 272

Notice: Undefined index: HTTP_REFERER in /var/html/www/AI/wp-content/themes/oxides-child/functions.php on line 272

Notice: Undefined index: HTTP_REFERER in /var/html/www/AI/wp-content/themes/oxides-child/functions.php on line 272

Chat With Us