UN urges moratorium on some AI systems

The United Nations Human Rights Office on Wednesday called on countries to halt the sales and use of artificial intelligence (AI) technologies that curtail human rights until measures are in place to prevent violations. 

It singled out AI systems that use facial recognition to rank people based on behavior, or group people based on gender or their ethnic background.

In a press conference, UN High Commissioner for Human Rights Michelle Bachelet warned of "catastrophic events" if the impacts of AI are disregarded, and called to introduce "guardrails" to prevent rights violations.

"The power of AI to serve people is undeniable. But so is AI's ability to feed human rights violations at an enormous scale with virtually no visibility," she said.

Facial recognition programs leading to profiling?

The press conference and a concurrent 16-page report released by the UN detailed how the reliance of countries and businesses on AI technology has led to unjust consequences.

In one example, the UN said innocent people had been arrested and identified as terrorists or fraudsters due to faulty facial recognition technology.

People had also been unfairly denied social security benefits due to faulty datasets and reliance on AI for decision-making.

 "The risk of discrimination linked to AI-driven decisions — decisions that can change, define or damage human lives — is all too real," Bachelet said.

UN says governments and businesses alike too hasty

According to the UN, businesses and governments alike have often hastily implemented AI technologies without properly evaluating how they work and what impact they will have.

As machine learning (more commonly referred to as "AI" despite being a small subset of that field) continues to grow and its use spreads, there needs to be more accountability in how data is collected, stored, and shared, the UN's human rights chief said.

Bachelet said that applications like government "social scoring" systems — used to monitor and judge people based on their behavior —  and certain AI-based tools that categorize people into clusters such as by ethnicity or gender should be prohibited.

"Action is needed now to put human rights guardrails on the use of AI, for the good of all of us," Bachelet said.

go/msh (AP, AFP)