Can Artificial Intelligence Replicate Human Decision Making?

The end of the 20th century laid the foundation of the information technology sector. In the coming years, there have been remarkable developments in the field. Today, we are at the cusp of the next big improvement in the digital technology sector: artificial intelligence. The IT revolution of the late 20th century enabled people to have quick and seamless access to information, but the ability of acting on the information laid predominantly on the operator. However, AI is reverting the status quo by automating the process of decision making, with the promise of improved efficiency.

However, Artificial Intelligence makes decisions that are based entirely on logic, and does not consider other factors whereas human decisions may be guided by a set of explicit rules, or by associations simply based on consequentialism, or by a combination. Furthermore, unlike machines, humans comprehend information selectively to make decisions. Due to the lack of selectivity, AI-powered machines may consider several irrelevant information while making decisions.

ethics.jpg

A prime example of this situation is when a computer program used by the United States Court of Law flagged a higher proportion of black ex-convicts (45%) as likely to become repeat offenders, as opposed to white convicts (23%). If such an act was reciprocated by a human, it would be considered an act of racism. This raises the question of whether AI makes decisions that are against human ethical values and whether the process of decision-making can be entrusted with machines that learn solely from past data.

As algorithms increasingly rely on features for improving predictability, the logic governing such decisions becomes more unconceivable. Subsequently, we lose the holistic aspect of decision-making, ignoring all ethics in favour of past observations. In some instances, this may be unethical, and in some, illegal results. Today, our sense of ethics provides a framework for making decisions. It may not be long before our decisions cast doubt on our ethics altogether.

 

 

EthicsVinay Gadagoju