Dear colleagues in Jordan and abroad
In my message of 2021, I did allude to 2021 as the year of big data being collected and managed as a resulted on the pandemic and of course the proliferation of information technology in health I general and artificial intelligence (AI) in particular. In that message I said “AI has ethical and legal implications. It would be wrong to assume that having an AI system will just run regardless of the culture, the community, the language, the legislative framework, and the value system in which it’s functioning”. In this message I wanted to focus on the ethics of AI using the data that has been collected and the extreme bias drives systems developers, intentionally or unintentionally to cause unethical practices in using AI. I wanted to confirm few facts: First of all, I would like to emphasise the AI is one form of eHealth or what is now called digital health; the second is that humans are to blame for writing computer codes that have inherited unethical practices and then blame algorithms for this unethical practices based on data or questions that people do collect or ask; third it’s important to differentiate between what is ethical and what is legal. There are examples of unethical practices related to the use of AI including racial discrimination in crime prediction. Members of the population who look “different” from the average are more likely to be suspected of committing crimes and predicting the likelihood of re-offending (repeated criminal behaviour) by arrested individuals. Another source of bias using the AI is gender discrimination in recruitment as the case of using algorithm to evaluate and rank the CVs of potential candidates to find new top talent resulting having a system that was not evaluating applicants in a gender-neutral way. The other source of bias is the stereotypes in the social media that translate into algorithm that creates bias towards or against certain race, national, ethnic origin, colour, religion, age, sex, family status, disability, and others.
Jordan has takena number of steps to create awareness of the importance of adopting ethical values that could be used in industry, commerce, public administration, and health. The national association for code of ethics was launched on 24 September 2006. The Ministry of Digital Economy and Entrepreneurship issued the National Charter of Ethics for Artificial Intelligence in 2022. The Charter calls for Jordan to develop a national strategy and regulatory regime for the deployment of AI technology. The deployment of AI will be done by “finding a common ethical base” based on “human and religious values, and the customs and traditions of society.” The charter contains several ethical principles for AI, including accountability, transparency, and respect for individual privacy, especially health data of individuals. The huge risk stems from the fact that many of the AI and machine learning software in the country are imported. This creates a challenge, in health, as healthcare is very much affected by the national value system. What might be accepted in another country, based on data that was collected in that country, might not be accepted in Jordan. Development of national AI algorithms or their adoption based on the cultural and religious values in the society is of prime importance to limit bias and possibly eliminate it.
eHDA in general and its President are supportive of ethical application of AI in the health and medical care sector in Jordan.
Sincerely yours
Najeeb Al-Shorbaji, President eHealth Development Association