Page 40 - Third Issue

“ enforcement loopholes. openness in data consumption as a

Privacy Laws and Regulations:
fundamental element. AI systems
must notify users when their data
In a nutshell, privacy rules and
The current privacy laws and rules
that regulate the usage of AI, such
as the General Data Protection regulations are crucial in the age is being gathered, processed, and
used as well as the reason for such
of AI for safeguarding individual
Regulation (GDPR) in the European privacy. These laws, which aim to actions, among other information.
Union. strike a balance between innovation For users to make educated
and protection, include the GDPR decisions and for data collection
An increasing number of nations and CCPA. While regulations are a and processing to continue to
and areas have created or amended positive beginning, their efficacy be transparent, this includes
privacy laws and regulations in depends on how well they are explaining how AI algorithms
response to the growing worries applied and enforced. Therefore, operate, including any biases or
about the potential for privacy governments must continue to discriminatory elements that may
abuses related to AI. The General set moral standards and laws that emerge.
Data Protection Regulation (GDPR), uphold people’s right to privacy and
which was put into effect in the encourage the responsible use of AI In this regard, acquiring meaningful
European Union in 2018, is one while also being attentive in their permission in the context of
illustration of this. oversight of AI technologies. AI-driven data collection and

The GDPR gives people more “ processing can be difficult, but it is
crucial to guarantee that people’s
Making Decisions with Information:
control over their data and aims to Difficulties in obtaining meaningful privacy rights are upheld. Crucial
preserve their privacy by regulating consent when data collection elements of this approach include
how it is obtained, used, and stored and processing are driven by transparent data usage and well-
by corporations. Strict rules for data AI. Examine the significance of informed decision-making. To
protection have been implemented making informed decisions and the ensure that people can make
by the GDPR, including data openness of data usage. informed decisions about their
minimization, informed permission, data and are aware of how their
the ability to access and delete It can be difficult to obtain data is being used, policymakers
personal data, and duties for meaningful consent in the context and organizations must put strong
data breach notification. Strong of AI-driven data gathering and controls in place.
financial penalties for non- processing. Due to the complexity
Techniques for Anonymizing
compliance with the GDPR’s rules of data processing, it is frequently “
serve as tremendous incentives to difficult for people to understand and De-identifying Personal
safeguard personal data and assure how their data will be gathered, Data: Examine the methods for
compliance. processed, and used. As a result, anonymizing and de-identifying
traditional permission models are personal data and talk about how
Although the GDPR has been frequently insufficient for AI-driven well they protect privacy in AI
praised as a big step towards data collection and processing. applications.
defending privacy rights in the Furthermore, people could not
age of AI, it is still unclear how be given the chance to give their By limiting the re-identification of
successful it will be in actuality. explicit consent or might not be personal data, anonymization, and
Regardless of where they are completely informed about how de-identification techniques are
located, all firms processing data their data will be used. essential for preserving privacy in AI
of EU residents must comply with applications. Anonymization is the
the rule, but doing so is still difficult Gaining meaningful consent process of deleting all identifying
and complex. Furthermore, because requires transparent data utilization information from a dataset to make
the law is still relatively young, and informed decision-making. it impossible to distinguish between
it is too early to judge its overall People must have access to different people. On the other hand,
effectiveness. The use of AI is also information that is transparent and de-identification entails eliminating
governed by different privacy laws easy to understand about how their particular identifiers or traits from
and rules throughout the world, data will be gathered, handled, and a dataset, making it more difficult
such as the California Consumer used. This data must be presented (but still possible) to identify specific
Privacy Act (CCPA) in the US. The in a clear and intelligible way. persons.
CCPA places new requirements Additionally, people should have
on businesses that gather or the option to refuse some data Personal data can be anonymized
handle personal data and gives processing activities with which and de-identified using a variety
Californian customers extensive they disagree or which they fear of methods, such as randomness,
rights over their data. However, may compromise the security of aggregation, and generalization.
similar to the GDPR, there are still their personal information. To hide identifying information
questions about its efficacy due to In addition, AI systems must include from a dataset, randomization
entails injecting random noise.
implementation issues and possible
40 ICT NEPAL (Issue-3, 2023)