How AI is already reshaping working conditions

This article is published in association with United Nations.

From delivery couriers compelled to follow the demands of online platform algorithms to content moderators who confront pornography and death every day while training artificial intelligence (AI) systems, the impact of new technologies on working conditions is becoming increasingly obvious.

Experts from around the world shared a range of concerns about how AI is affecting work and presented new ways to harness new technologies that protect rights and foster prosperity during a recent webinar organised by the International Labour Organization (ILO) and the International Telecommunication Union (ITU).

“The key issue is not whether AI will transform work; it already is,” said Sher Verick, ILO coordinator for digitalisation and AI. “The central issue is how to ensure that this transformation advances decent work and social justice.”

At the first meeting of the Independent International Scientific Panel on AI, held on Tuesday, UN Secretary-General António Guterres emphasised the new body’s “huge responsibility” in “helping shape the trajectory of artificial intelligence for the benefit of humanity”. 

“Individually, you come from diverse regions and disciplines, bringing outstanding expertise in AI and related fields,” the UN chief said. “Collectively, you represent something the world has never seen before.”

Making AI ‘trustworthy’

Artificial intelligence is now being used in systems “with real consequences for people’s prosperity”, said Bilel Jamoussi, Deputy to the Director of the ITU Telecommunication Standardisation Bureau.

That includes hiring decisions and access to services. He emphasised the importance of standards and skills development, noting that technical standards help make AI “trustworthy”, but that the decisive factor remains how societies and governments choose to apply these technologies.

Meet the ‘invisible’ workforce

An “invisible” workforce that keeps AI running stands behind chatbot responses, social media algorithms and other automated systems that many people now take for granted. 

During the webinar, Ben Richards of UNI Global Union explained that “when we speak about workers in the data supply chain, we’re talking about two main groups:”

• content moderators that keep platforms safer by reviewing harmful content
• data labelers and annotators who structure the reality so machines can learn

Wherever his organization speaks to such workers, he added, they describe similar conditions: “extreme pressure, constant monitoring, low wages and mental health harms”.

Trauma behind the screens

Major technology companies often rely on workers in countries of the Global South for content moderation and data annotation. 

In India alone, tens of thousands of people are engaged in such work. For many rural residents, particularly women, job advertisements offering work from home with only an internet connection required appear to be a rare opportunity for income and independence.

When they are hired, however, they often do not know what material they will be expected to review. Many are required to sign non-disclosure agreements (NDAs) prohibiting them from discussing their work even with family members.

Sexual violence and traffic accidents

A woman from a village in India described watching hundreds of videos per day, including scenes of sexual violence, traffic accidents and people dying, according to a recent media report.

Another young woman said she was required to review content involving sexual violence against children and classify pornographic material.

Human rights advocates have raised concerns about such working conditions.

© Unsplash/Finn

AI technologies are changing the landscape for workers around the world.

Global alliance promotes safe-work protocols

Mr. Richards of UNI Global Union noted that his organization is building a global alliance of content moderators and promoting safe-work protocols grounded in the right to organise and engage in collective bargaining.

“We want AI to augment human capacities,” he said.

Indeed, the benefits of technological progress must be distributed fairly, he stressed.

Chasing impossible targets

Workers from across countries are also feeling increased stress from algorithmic management, in which software systems determine work pace, task allocation and performance evaluation. This type of management can intensify pressure on workers and even create safety risks.

Two thirds of drivers and couriers in the United Kingdom work under conditions of anxiety due to “unfair feedback” and sudden changes in working hours determined by algorithms.

Evelyn Astor, director of economic and social policy at the International Trade Union Confederation (ITUC), warned that without appropriate regulation, artificial intelligence could deepen existing risks.

According to trade union monitoring, there have already been fatal accidents linked to couriers “chasing impossible delivery targets set by algorithms”.

Around two thirds of drivers and couriers in the United Kingdom work under conditions of anxiety due to “unfair feedback” and sudden changes in working hours determined by algorithms, according to a 2025 study published by the University of Cambridge. More than half of respondents said they risk their health and safety at work.

Although platforms do not explicitly instruct workers to violate safety rules, the system of incentives, including penalties, speed-based bonuses and priority order allocation, creates conditions in which workers feel compelled to make dangerous decisions in order to preserve their income.

Shaping AI’s trajectory to benefit humanity

Similar concerns are emerging in other countries and sectors: automated systems can assign shifts, set pay levels or even make termination decisions, often with minimal human oversight and limited avenues for appeal.

Trade union representatives at the webinar warned that deploying AI without proper safeguards risks reinforcing existing problems, adding that algorithmic management is already contributing to work intensification, increased digital surveillance, biased decision-making and reduced human oversight in employment decisions.

The ILO and ITU are advancing initiatives aimed at shaping international approaches to regulating AI’s impact on labour markets, including the AI for Good platform and the Global Coalition for Social Justice.

For the United Nations system, the central challenge is to ensure that artificial intelligence expands human potential rather than undermines workers’ safety and wellbeing, which requires shifting the focus from technological innovation alone to governance grounded in human rights, equality and sustainable development.