Top 10 jobs at risk as artificial intelligence threantens 40% of jobs worldwide;see full list

The International Monetary Fund (IMF) has warned that AI could affect about 40 per cent of jobs worldwide, as the technology reshapes how businesses operate and how workers perform their duties.

According to a report published by Citi Newsroom on October 12, 2026, IMF Managing Director Kristalina Georgieva delivered the warning during a panel discussion at the 2026 IMF and World Bank Annual Meetings in Bangkok, Thailand. She cautioned that AI could transform jobs, reduce the value of some roles or make certain positions redundant.

The IMF estimates that AI could affect around 60 per cent of jobs in advanced economies, compared with 23 per cent in low-income countries. The impact will depend on how widely the technology is adopted and the types of tasks involved.

While AI is expected to create new opportunities and improve productivity in some sectors, workers whose duties can be automated may face growing pressure to adapt.

Here are 10 jobs that could face significant disruption as AI becomes more widely used. This is not a definitive list of jobs that will disappear; the technology may automate only some tasks while changing the responsibilities of workers.

Data entry clerks could face increasing pressure as AI-powered tools become better at extracting, organising and transferring information between digital systems.

Businesses can use automation to process forms, update databases, digitise documents and identify errors with less manual input. As these systems improve, some organisations may need fewer workers dedicated solely to entering information.

However, employees who can manage data quality, resolve complex errors and oversee automated systems may remain valuable.

Customer service is another area where AI could reduce demand for some routine roles. Chatbots and virtual assistants can answer frequently asked questions, provide basic product information and help customers track orders or resolve straightforward complaints.

Companies may increasingly rely on these systems to provide round-the-clock support, reducing the volume of calls and messages handled by human agents.

Human representatives are still likely to be needed for sensitive complaints, complicated cases and situations that require empathy and judgement.

AI-powered voice assistants can make outbound calls, respond to common enquiries and handle simple customer interactions. Some businesses may use them to follow up on leads, confirm appointments and conduct basic surveys.

This could affect telemarketing and call centre jobs that involve following scripts and repeating the same information to large numbers of people.

The extent of the disruption will depend on customer preferences, business needs and whether automated systems can handle conversations effectively without human intervention.

Administrative workers who primarily schedule appointments, organise records, prepare routine correspondence and manage basic office tasks could see parts of their work automated.

AI tools can draft emails, summarise documents, arrange meetings and retrieve information from digital files. These capabilities may allow smaller teams to complete work that previously required more administrative support.

Nevertheless, assistants who coordinate complex activities, manage confidential information and support decision-making may continue to play an important role.

Transcriptionists convert recorded speech into written text, a task increasingly handled by speech-recognition software.

AI tools can transcribe interviews, meetings, lectures and recordings in a fraction of the time required for manual transcription. Some can also identify speakers, generate summaries and organise transcripts by topic.

This could reduce demand for straightforward transcription services. Human professionals may still be needed to check accuracy, interpret unclear audio and handle specialised material involving legal, medical or technical terminology.

AI writing tools can identify grammatical mistakes, suggest alternative wording, improve sentence structure and adapt text to different tones.

These capabilities may reduce the need for human intervention in basic proofreading and routine editing, particularly where the material is straightforward and does not require specialist knowledge.

However, editing involves more than correcting grammar. Human editors remain important for checking facts, assessing context, preserving an author's voice, identifying misleading claims and making editorial judgements that automated systems may get wrong.

AI image-generation and design tools can produce social media graphics, simple advertisements, presentation layouts and draft visual concepts from written prompts.

Small businesses and individuals may increasingly use these tools for routine designs that once required a professional to create from scratch.

This could put pressure on designers whose work focuses on repetitive templates and simple visual assets. Designers with strong creative direction, branding, art direction and client communication skills may be better positioned to adapt.

AI-enabled accounting software can categorise transactions, process invoices, reconcile accounts and flag unusual entries. These functions may reduce the time required for routine financial administration.

Businesses could therefore automate some tasks traditionally performed by bookkeeping and accounting clerks, particularly where transactions follow predictable patterns.

Human professionals will still be important for interpreting financial information, investigating discrepancies, ensuring compliance and advising businesses on financial decisions.

AI can search large volumes of documents, summarise reports, extract relevant information and help organise research findings. This could affect entry-level positions that primarily involve gathering information and preparing initial summaries.

The risk is particularly relevant to early-career workers who rely on these assignments to gain experience before progressing to more specialised responsibilities.

However, research still requires people to assess source credibility, verify claims, identify gaps in evidence and draw sound conclusions. Workers who combine AI tools with critical thinking and subject knowledge may remain in demand.

AI translation tools can convert text between languages quickly, making them useful for routine documents, general correspondence and basic online content.

As the technology improves, demand may fall for some straightforward translation assignments, especially where a high level of cultural nuance or specialist accuracy is not required.

Human translators will continue to be important for literary work, legal documents, sensitive communications and material where local expressions, cultural meaning and context are essential.

The IMF's warning does not mean that 40 per cent of jobs worldwide will automatically disappear. The estimate refers to jobs that could be affected by AI, including roles that may be enhanced by the technology, changed in value or eliminated.

The effects are also unlikely to be evenly distributed. Some workers may become more productive with AI assistance, while others could face reduced hiring opportunities or pressure to develop new skills.

Young people entering the labour market may be particularly exposed if employers automate routine entry-level assignments that traditionally help graduates gain workplace experience.

For workers and businesses, the challenge will be to identify which tasks can be automated and which still require human judgement, creativity, accountability and interpersonal skills.

The IMF has called for investment in digital infrastructure, workforce training and education reforms to help countries prepare for the transition. For employees, developing digital literacy, learning to use AI responsibly and strengthening skills that complement automation could become increasingly important.

Ultimately, AI is likely to change many jobs rather than simply eliminate them. How disruptive that change becomes will depend on the pace of adoption, the decisions employers make and how effectively workers and governments prepare.