Artificial intelligence is expanding rapidly at an astonishing pace raising concerns about its use,
ownership, accountability, transparency, privacy, security etc. As AI has the potential to surpass
human capabilities in the present, the need to address the ethical challenges involved is of
paramount importance. It has to be scrutinized at each step so as to harness the AI’s immense
potential.
Ethical issues related to Artificial intelligence in the present scenario:
Transparency and accountability:
AI works in a closed circuit method wherein the interpretation of the work done by AI cannot be
tracked and ascertained. This is especially required in areas of research in healthcare domain,
transparency must be ensured to improve the reliability of the work done by AI. Accountability
is required if any errors have occurred by chance, necessary corrective actions must be taken.
Bias and discrimination:
AI systems can possess bias in algorithms as it has a huge reservoir of data embedded in it.
These bias can be perpetuated and amplified resulting in discriminatory outcomes in crucial
areas of research. Ai system can reproduce or amplify racial, ethnic, gender, political and other
biases in the training and subsequent data received. Stringent protocol must be developed to
combat bias in the AI system.
Creativity and ownership:
A new research work can be created by a researcher in an AI system. The onus of the research
work created by the researcher must be given to the researcher and not the AI system
organization. Infringement laws must be clearly stated in using AI systems.
Social manipulation and misinformation:
The research ideas and data available can be manipulated and false, fabricated information can
be spread using AI algorithms. Vigilance and countermeasures have to be taken to manage this
issue.
Data privacy:
Personal data must be protected while using AI systems. The data stored in AI systems should
comply with privacy regulations and there must be robust measures to secure the data safely.
Informed consent:
Obtaining informed consent from participants when collecting data must be used to train AI
models. The participants must be clearly explained how their data will be used for the purpose of
research.
Societal impact:
AI has the capability to displace personnel from their jobs in areas of research. This can hamper
job prospects available in research areas like data collection, data handling etc.
Continuous monitoring and auditing:
AI systems must be continuously monitored for any bias or issues and corrective actions must be
taken.
Judicious use of AI in research:
Researchers must disclose and clearly explain the way AI has been used in their research. They
should also specify the limitations of using AI in their research.