How to solve the challenges in adopting Artificial Intelligence

Drawing on an AI screen

Written by Megana Natarajan, Global News Correspondent with OnDOt Media

CIOs acknowledge that AI was treated as an all-in-one solution by most organizations, but there are several obstacles in its implementation in any industry

AI has been rapidly embraced by organizations, which has given rise to the dialogue on how the platform needs to be further developed and deployed. It has played a major role in helping firms tackle cybercrimes and also helped the healthcare industry to provide a faster response during the pandemic.

However, IT leaders say that an organization must be well aware of AI’s shortcomings before implementing it in the system.

 

Lack of ethics and governance policies

IT leaders acknowledge that AI implementation is largely done without a set of ethics guidelines. A structure is imperative to ensure effective implementation along with governance policies and ethical principles.

CIOs point out that AI tech is built by humans with human prejudices and ideas. There have been examples where discriminatory standards were unintentionally fed into the system due to the standards followed by the developers.

Case in point being the facial recognition tech, run by AI, and was unintentionally biased against the Black and other minority communities. Such mistakes happen due to unconscious bias present in the data set fed into the system.

IT leaders say that these issues can be solved by creating a governance protocol that prioritizes ethics, builds transparent, descriptive, and robust systems with a clear audit trail and models which can adjust to changes, and finally measured updates and roll-outs with clearly documented processes.

 

Lack of AI talent and training

IT leaders think that while AI is the most transformative tech of the 21st century, its deployment needs very clear and strong methodologies.

Organizations should offer relevant training to its employees to ensure effective adoption of the technology. This will ensure higher productivity form AI usage since when employees are empowered and understand the tech, they can use it better and even upskill themselves.

Security professionals point out that AI can be easily misled and used for malicious activities. The decision-making skills of AI are largely still considered to be black-box and its liabilities a cause of concern. Today, IT Security leaders say that organizations are recording increased cyber-attack incidents, especially against their AI systems.

CIOs believe that the best solution to this is transparency. Strategic measures that can document AI’s decision-making protocols are required to be in place. This is critical to establish trust and thus increase the adoption of the tech.

Organizations also need to ensure that they practice infrastructure and data readiness before deploying AI tech. IT leaders say that this can be achieved via strong data strategies, cleaning, validation, organized data, and standardized methods implemented across an enterprise. Infrastructure investment is also critical to support AI solutions as computing power is pertinent to enable AI.


Originally posted here

More thought leadership

Join Digital Leaders

By submitting your contact information, you agree that Digital Leaders may contact you regarding relevant content and events.
Analytics by GoSquared