Science & Tech
JIDE Awe, a Science, Technology and Innovation (STI) Policy Advisor, on Sunday stressed that privacy and data protection are key ethical considerations for the influx of Artificial Intelligence (AI) into the country.
In an interview with the News Agency of Nigeria (NAN) in Lagos, Awe said that one reason necessitating increased attention to data protection and privacy was the immense reliance of AI systems on data.
He said that data must be collected, used, processed and stored ethically to protect user privacy and respect their privacy rights.
Awe, who is also the Founder, Jidaw.com, said that bias could be problematic in AI.
According to him, in order to ensure AI is not used to foster and entrench unfairness, discrimination, and oppression, it is ethically responsible to identify and mitigate bias in the data that trains AI systems.
“Other important ethical considerations in use of AI are responsibility and accountability; What is the responsibility of developers? How can the deployer, project owner or the user be held accountable for the actions of AI systems?
“Though Nigeria currently lacks a comprehensive legal framework for AI, data protection regulations are essential in line with the Nigeria Data Protection Act (NDPA).
“Regulation should enable impact assessment to evaluate the potential social and ethical impacts of AI systems before and after deployment,’’ he said.
According to him, human oversight is also crucial as AI systems are programmed machines that can not think for themselves.
Awe noted that there was a need to ensure AI-made decisions were guided by common sense and critical thinking.
According to him, unequal access to AI technology can also exacerbate existing inequalities, adding that inclusive AI development that benefits all Nigerians is crucial.
The STI advisor said that transparency in AI decisionmaking was crucial, instead of the black box approach.
According to him, it is important to know AI systems reach decisions which make human oversight and intervention possible.
Awe said: “Regulation on AI should be efficient, effective, and flexible to adapt to the changing circumstances of our technology-driven world and to foster innovation.
“Having standards and certification processes for AI systems could promote responsible development and use.
“Regulation should also promote an environment of continuous learning and ensure standards are kept up to date.’’
Awe described the country’s preparedness towards AI as unsurprisingly a combination of positives and challenges.
He said that the Federal Ministry of Communications, Innovation, and Digital Economy had started the AI readiness process with the development of a National AI Strategy, which began with an AI Experts workshop.
Awe said that this workshop led to the launch of Nigeria’s first Multilingual Large Language Model (LLM).
According to him, Nigeria also collaborates in international AI development efforts and is a signatory to relevant declarations.
He added that there were some AI development and advocacy efforts, along with capacity building and talent training programmes in Nigeria.
The STI expert, however, noted that the level of AI-readiness activity remained low with minimal impact, saying that it was also primarily driven by private research initiatives.
“Unreliable internet, limited computing power and inconsistent public electricity, coupled with inadequate talent and research capabilities are major obstacles to AI development in the country,’’ Awe stressed. (NAN)
F.A
July 1, 2024
Tags: DataProtection
