Zulkhadir Riandi
Feminist Society - Today, the course of life in the era of the 5.0 industrial revolution has brought increasingly heavy digital and technological developments, in which digital and technological advances continue to evolve. One of the connections is with the existence of Artificial Intelligence (AI) that continues to advance and evolve throughout all lines of human life. Where today, almost anything in the digital world can be run easily with the help of AI.
On the one hand, AI has a positive impact because it can
make it easier for us where AI can generate whatever we order based on clear
and specific prompts. But on the other hand, AI actually has a negative side,
where AI tends to be gender-biased. For example, AI or other search engine mechanisms
tends to generate or deliver results with a majority of men in jobs such as
doctors and women in nurses. Another example can be seen when we carry out
searches related to everything related to childcare, which is always associated
with the mother, as if it were only the role of the mother, whereas childcare
is the responsibility of both parents (mother and father).
You can also Read: The Importance of Paternity Leave in Promoting Gender Equality and Family Well-Being
AI systems are biased because they are human creations. Who
makes decisions informing AI systems and who is on the team developing AI
systems shapes their development. Unsurprisingly, there is a huge gender gap:
Only 22 percent of professionals in AI and data science fields are women—and
they are more likely to occupy jobs associated with less status. In terms of
gender bias from data, data points are snapshots of the world we live in, and
the large gender data gaps we see are partly due to the gender digital divide.
For example, some 300 million fewer women than men access the Internet on a
mobile phone, and women in low- and middle-income countries are 20 percent less
likely than men to own a smartphone. These technologies generate data about
their users, so the fact that women have less access to them inherently skews
datasets. Even when data is generated, humans collecting data decide what to
collect and how.
How to Mitigate Gender Bias in AI?
Mitigating gender bias in AI algorithms is crucial for
creating fair and equitable systems. Here are some strategies:
1.
Diverse Data Collection
Gathering diverse and representative data during model
training will have direct implications in overcoming gender bias in AI. Therefore,
we have to ensure that the dataset includes various gender identities,
backgrounds, and experiences.
2.
Bias Detection and Audits
Regularly audit AI models for bias. Identify discriminatory
patterns and adjust the algorithms accordingly. Tools like fairness metrics and
adversarial testing can help. This mechanism is of paramount importance to
persistently prevent gender bias in AI systems.
3.
Feature Engineering
Be mindful of features that might introduce bias. Remove or
adjust features related to gender, race, or other sensitive attributes.
Thereby, the full awareness of the technicians and all parties involved in the
creation of AI systems is really needed.
4.
Balanced Representation
Oversample underrepresented groups to balance the dataset.
This helps prevent the majority group from dominating the model’s predictions.
The more representative of gender participation in creating AI systems is, the
better AI systems will be.
5.
Ethical Guidelines
Develop clear guidelines for AI development that address
bias. Involve ethicists, social scientists, and affected communities in the
process. By making good guidelines for AI creation, it will drive a better
procedure that takes into consideration gender equality.
Remember that bias elimination is an ongoing effort. By
combining technical solutions with ethical considerations, we can create AI
systems that treat everyone fairly.
If you are interested in issues related to gender equality
and in-depth discussion about feminism, let's join with Feminist Society.
Feminist Society aims to promote progress in gender equality and women’s rights
through education and advocacy. To get related articles, please visit the page
feminist-society.blogspot.com.
You can also contact us on:
Instagram :
@feministsociety.id
Email :
fem.official7@gmail.com
View more:
Ayesha Nadeem et al. Gender Bias in AI: A Review of
Contributing Factors and Mitigating Strategies. AIS Electronic Library. (2020).
https://aisel.aisnet.org/acis2020/27/
https://analyticsindiamag.com/understanding-ai-biases-and-ways-to-fix-them/
No comments:
Post a Comment