When AI harms humans, who is to blame? Global experts grapple with accountability

“In 11 Southern countries, one child per class reported that AI was used to create sexually explicit deepfakes of those children. AI-generated child sexual abuse material to CyberTip’s US channel, NCMEC, is increasing exponentially,” said Sonia Livingstone, of Independent International Global Scientific Panel on AI.

Highlighting other key findings from the panel’s first preliminary report mandated by the UN General Assembly, Ms Livingstone stressed that the evidence of AI-related abuses against individuals, vulnerable and disadvantaged groups is “far more compelling” than its benefits.

This is because of the technology used”to create and strengthen persuasion and deception, spread disinformation, distrust”with personal data “taken, manipulated, misused, exploited” by self-learning systems, he stressed.

Universal problem

It is precisely because AI can be used for harm – and for good – that the UN believes that all countries should be involved in deciding how the technology should be managed.

The first Global AI Dialogue in Geneva – on the sidelines of the Global AI Conference AI for Good Summit organized by the United Nations International Telecommunication Union (ITU) – will be followed by a second summit next May in New York.

One way to ensure that AI benefits humanity rather than harming it is to embed equality, accountability and human oversight in its design, development and implementation, he stressed. Volker Türk, UN High Commissioner for Human Rights.

“We do not consider safety standards for medicines, cars or airplanes as a barrier to progress. These are the reasons why people trust this technology,” he stressed.

Power struggle

Among the many unresolved governance issues, one of the biggest is the energy-hungry data centers that power today’s AI platforms – prompting calls for industry-wide transparency from the United States.N Secretary General.

“Anyone with a cell phone, with cellular service, a laptop can connect to the cloud and benefit from AI to some extent. The impact, the negative impact that we see is very, very local,” said Sasha Luccioni, co-founder of the Sustainable AI Group.

Some of the world’s most marginalized groups are already suffering “in terms of water, energy and emissions, as well as worsening health impacts”, he said.

In addition to the challenges of deciding how to measure AI’s impact on the environment, digital policy expert Jhalak Kakkar, from the Center for Communications Governance at Delhi’s National Law University, is worried about this. risks exacerbating existing inequalities in the countries of the South.

“How do we deal with the implications of economic and other value derived in locations that are very, very far from the communities most affected?”.

Code blocks

It is also increasingly clear that as AI advances, it becomes increasingly difficult to maintain checks and balances at the legislative level.

“We must be able to open the black box of AIstressed Amal El Fallah Segrouchni from Morocco, who explained that large language models utilize around 180 billion parameters while operating 24 hours a day, seven days a week. “There are very few lawyers and very few sociologists who are in a position to understand exactly what is going on,” he said.

“Even AI researchers sometimes don’t have the tools to understand the complexity of these algorithms.”

Online abuse trends

Currently, one in four women human rights defenders, activists and journalists take part in a survey conducted by the UN gender equality agency, UN Womensaid they had experienced AI-assisted online violence; six percent said they had been victims of deepfakes or digitally manipulated images. “A lot of this is undocumented, unknown, unreported,” it said Sima Bahous, Executive Director at UN Women.

Related to this is the fact that “when we look at who is building AI, women are largely underrepresented” and make up only 30 percent of the global workforce in tech, the agency head stressed. According to the report of the Independent International Scientific Panel on AI, 88 percent of leading AI researchers are men.

Universal debate

That Global Dialogue on AI Governance exists to ensure that the priorities of all countries are included and not just the most technologically advanced countries. The UN firmly believes that the benefits of AI should be felt by everyone.