AI has many positive impacts. That’s why it is being rapidly integrated within our daily lives; whether it is detecting potential cancer risk, facial recognition of traffic offenders or deciding what content we see on our social media, AI is a part of all the processes. Therefore, regulating AI and its potential risks and dangers is extremely important.
Now, how can AI pose dangers?
Well, for starters AI uses a huge amount of datasets for training and development. This means that the companies that have to use AI need user information for creating algorithms. This poses a threat to the privacy of users. Secondly, the dataset that AI analyses can be biased. This can cause discrimination to increase against already less represented groups like women, people of color and other minority groups. The way AI is trained, can cause it to make more errors and also increase inequality. For example, the content posted by a minority group can be suppressed by AI hence, leading to less diversity on the digital platform. Also, the working of AI systems is opaque, ie, users don’t know how it works and therefore dont know its decision making process which creates trust issues in the information provided by ai. These reasons make monitoring of AI extremely important.
So, how can we protect human rights and advocate problems regarding AI?
Designing laws and regulations specific to AI is the critical need of today. As of now, most countries don’t have specific laws regarding ai. The UN has been constantly working and trying to frame laws specific to ai. Recently, in 2024, the UN and EU passed the ai act which promotes safe,secure and trustworthy ai. Many companies follow the FAT principles(fairness,transparency and accountability) when developing AI. This promotes accountability of designers and deployers, and transparent and inclusive AI designs. Other than this, using a risk-based approach towards designing AI can reduce potential risks. In this approach, designers need to think about the potential risk of Ai and create controls for these based on the amount of damage that they can do.
In conclusion, the creation of AI laws needs to be aligned with the rate at which AI is developing to ensure that AI poses to be more beneficial than harmful.