Every boardroom, every university lecture hall, and every kitchen table is poised for the same conversation. They are asking, “Is my job safe?” It is one of the most pressing questions of our time, and it merits an answer. We are not asking for a comforting or an alarming response. We simply expect the truth and the data to support this.
What will artificial intelligence do to the global workforce? Which jobs will vanish? Which jobs will become more available? Who is at risk? Who is safe? What can be done about this? This guide will answer all of these questions for you.
The Important Numbers
What is the research actually saying? Let’s look past the headlines. Let’s analyze the data from the World Economic Forum.
The Future of Jobs Report interviewed over 1,000 employers from 55 different economies. The report states that by 2030, the combination of AI and other technologies will generate 170 million new jobs around the world and eliminate 92 million jobs. They claim this will be a net creation of 78 million jobs. On paper, this report states some positive news.
But here’s an underreported story: destroyed jobs and created jobs aren’t the same. They’re not the same skills or wages, and they’re not even in the same places. A small town’s former customer service representative does not transform into an AI engineer in San Francisco. The true story exists in this disparity, who loses jobs and who gains jobs.
According to McKinsey, there is now, and even in the future, the capability to automate more than half of the working hours in the U.S. with current technologies. It won’t be the case that 57% of jobs in the U.S. will disappear. Instead, AI will be able to perform the majority of the tasks that the working population will perform. The limit will no longer be what the technology can do, but the speed of deployment.
Goldman Sachs estimates that between the long-term and the short-term, approximately 6 to 7% of the U.S. labor force, around 11 million people, and approximately 300 million full time jobs in the world will be affected by generative AI.
What “Affected” Really Means
Here is where most reporting on this subject completely misses the mark. To say that someone has been “affected by AI” does not mean that they have been “replaced by AI.” Displacement means the job goes away completely and AI does it instead. Augmentation means that AI helps you do your job quicker and easier but you are still employed and are even more productive.
Transformation: tasks become automated, but new tasks emerge that require critical human judgment. Federal Reserve Bank of Dallas analysts predict that through 2026, wages in jobs most exposed to AI and machine learning will not decline. This suggests that for many workers, AI adds to their output rather than replaces it. This is a valuable distinction. The risk is thinking it will always be that way.
Which Jobs are Most Likely to be Affected?
The Honest List
All of the jobs that are most likely to be affected do the same type of work. These jobs are all highly predictable, rule-based, and most involve some iteration of a digital or text-based component. Jobs at the top of the list for the greatest risk in 2025 and 2026 include data entry, customer service, legal research, coding, and content moderation. These jobs are not low-skilled jobs. Many have historically had very good pay. These tasks involve work that AI and large language processing models excel at.
In finance, the numbers are equally alarming. The IMF states that 60% of jobs in developed countries will have a high exposure to gemini ai, with financial services at the highest end. It is estimated that Wall Street banks will aim to eliminate approximately 200,000 jobs within three to five years. Back office and entry-level jobs will be the most impacted. Up to 54% of banking jobs will be highly likely to be automated.
The reckoning in the legal profession is comparable. Paralegals face 80% job automation risks from gemini ai systems by 2026. Legal researchers face a 65% risk by 2027. AI doesn’t make better lawyers; AI systems can perform research and legal documentation at a fraction of the cost and replace junior legal staff functions. In automation, transcription is already 99% automated in the medical field, and 40% of medical coding is predicted to be automated by 2025.
Software engineering is a discipline reshaped by the very technology people entered the field for. Microsoft CEO Satya Nadella has stated that AI now writes 30% of the code Microsoft produces, and Dario Amodei, the CEO of Anthropic, has stated that AI could automate approximately half of the entry-level white-collar jobs in the next 5 years.
Lowest AI Displacement Risk
Jobs at the lowest risk for AI displacement share physical and unpredictable tasks, less automation, and hands-on skills. Electricians, plumbers, and HVAC craftsmen work in environments that require physical, real-time problem-solving that no robot handles well. Mental health counselors build real, empathetic therapeutic relationships that generative ai impact on jobs can’t. No AI can perform complex and highly unpredictable surgery. Senior creative directors judge in the ever-changing and evolving realm of aesthetics for customer/client needs.
According to the WEF, despite fears of automation, some of the least impacted areas, including farm work and construction work, have the most projected job growth for the new decade.
The White-Collar Surprise
The educated, white-collar workforce seems to be experiencing displacement in ways blue-collar workers are largely avoiding; this is perhaps the most shocking finding from the research conducted recently.
The University of Pennsylvania-OpenAI research collaboration has indicated that workers in knowledge-based jobs and earning salaries in the sub-$80,000 range are highly likely to experience the unfavorable employment effects of automation. This has never happened in the course of previous waves of automation. The Industrial Revolution was about the replacement of manual physical work. The machines of the industrial age built systems to automate manual labor. The AI systems presently being built and integrated into workplaces to replace white-collar workers are of an advanced and refined nature, and are capable of automating tasks of a cognitive nature.
Ford’s Jim Farley has predicted that generative ai impact on jobs will “replace literally half of all white-collar workers.” Salesforce’s Marc Benioff has claimed that AI is already accomplishing almost 50% of the work in his company. AI has caused a shift in the employment landscape so that fewer workers will be hired in the coming years, according to JPMorgan’s Jamie Dimon.
These are not predictions; these are modern-day realities being reported by those in charge of running the companies.
Research from Cornell University has indicated that AI adoption in the corporate sector has been accompanied by a 13% drop in the hiring of junior employees. The concern here goes beyond immediate survival. When there is a continuous decline in junior workers, there is a continuous decline in senior workers over time. The ladders of career progression are being removed.
The Gender Dimension Nobody Talks About
In the uneven gender distribution resulting from the displacement caused by AI, there is a dimension that should be addressed with much greater urgency.
In the US, about 79% of working women are in jobs considered highly threatened by automation, compared to 58% of men. This shows that the women in the workforce are in specific job classifications. For example, jobs that are predominantly female, such as secretarial jobs, medical administration, legal aid, and reception, are some of the jobs that are highly exposed to google ai.
Worldwide, the jobs that are highly exposed to AI disruption for women are 4.7%, and for men, they are 2.4%. If a discussion is to be held regarding the economic implications of AI, this will have to be included as part of the discussion.
What Is Actually Growing
This disruption is undeniable, but it is only one part of the story. The other part of the story, and a very real part also, is that jobs are being created.
In the first quarter of 2025, 35,445 google ai related jobs were posted in the US, increasing postings by 25% in the previous year. The median yearly salary for these jobs was almost $157,000. AI and data related jobs had 58,000 openings, and their median pay was $170,000. Data Scientists have a projected growth of 34% by BLS for the years 2024 to 2034.Computer Scientists and Research in IT have a 20% projected growth rate. According to the 2025 Global AI Jobs Barometer by PwC, the 2025 wage premium for employees performing AI jobs, compared to employees doing the same job sans AI, was 56%. The previous year’s premium was 25%. The premium is accelerating.
Jobs that utilize AI will continue to expand rapidly, increasing 3.5 times more than any other job sector. By 2030, McKinsey predicts that AI will have developed anywhere from 20 to 50 million new jobs globally. The LinkedIn Economic Graph noted that there were 1.3 million AI-related jobs and over 600,000 AI-related data center jobs between 2023 and 2024.
Cybersecurity is expanding rapidly as well. AI’s ability to create code has led to the creation of new bugs and vulnerabilities, leading to a shortage of security professionals to address the issues.
Newly Important Skills
According to the WEG, the most in-demand skill in 2025 will be analytical thinking. Seven out of ten companies agree. Skills such as these are not tool-based. They are centered around reasoning for problem-solving. AI can help with this, but the replacement of such reasoning is still not possible.
Employer surveys suggest companies are looking for employees who can generate AI outputs and use AI tools to their fullest potential. However, they are also searching for employees who understand the limitations of the tools, can spot mistakes, ask the proper inquiries, and make business decisions based on their findings. Skills that are in immense demand are data literacy, google ai prompt engineering, advanced problem framing, communication with ethical deliberation for AI, technical and domain expertise, and interpersonal skills.
Current projections suggest that 39% of skill sets will likely become obsolete or undergo major changes between 2025 and 2030. In response, 77% of employers will reskill or upskill their employees to ensure that their personnel will be able to work effectively with the AI tools available to them. Now is not the time to develop your skill set, especially given the level of disruption taking place.
Real Actions by Businesses
While statistics provide information, it is often abstract and difficult for most to understand. The Real Actions by Businesses section helps put the statistics into real-world, grounded understanding by illustrating what some of the world’s largest companies are actually doing.
The AI Investments being made by Workday have come at a cost. 8.5% of their workforce, about 1,750 employees, have been laid off. Leaner structures and faster innovation, according to Amazon, are the result of AI replacing the need for 14,000 corporate employees. In the first half of the year 2025, 77,999 employees have been laid off, due to various AI implementations making their positions redundant, that equates to about 491 layoffs every working day in the Tech Industry.
Though companies will rarely say it is their intention to “replace people with AI,” most have stated or claimed that they are restructuring or improving their operations in a strategic manner. The actual reason is that AI has begun taking over a lot of the functions that require human employees to perform, and AI doesn’t require physiological needs to be met.
What You Should Do
For Employees
The most dangerous response to disruption caused by AI is to do nothing. The second most dangerous is to respond with panic. The most productive path relies on a few, fairly straightforward principles. To begin, determine where your job currently falls with respect to the risk of automation. Those roles which center around the processing of information, the following of step-by-step instructions, or applying rules to repeatable inputs are more at risk compared to roles that involve unpredictable human behavior, some level of physical variability, or creative activities.
Next, develop a basic level of AI knowledge, no matter your occupation. This does not mean taking the time to learn computer programming. Rather, it means developing an understanding about how various AI tools function, their use cases, and where their limitations are. Those who are able to properly use AI tools are far more productive than those who are unable to, and this divide is only growing.
Additionally, develop the types of skills that actually work synergistically with AI as opposed to competing with it. These include advanced expertise in your field as no general AI system will be able to do that, and skills such as building trust, explaining complex ideas to those with little or no background in the relevant field, and deciding and acting correctly in a situation that is not very clear.
For Organizations
Companies that use AI to only automate processes and save on costs are making poor, shortsighted decisions that will ultimately impact their bottom line negatively. Stopping the recruitment of entry-level employees, which many companies are doing, will create an absence of employees with the required skills at the mid/senior management levels in 5-10 years.
85% of employers say that upskilling is their response to AI disruption’s number one workforce strategy. Organizations that invest in their human resources to adapt to the disruption of AI, while implementing AI to make their operations more efficient, will thrive in this era.
The Bigger Question
Some questions about the meaning of work can’t be answered by statistics. What do we want from work? is an example.
The relationship between technology and labor has changed in the ways that first seemed bad and then good. The best examples are the Industrial Revolution, the computerization of offices, and automated manufacturing. Each of these destroyed types of work but also created new ones.
There is good reason to suspect that AI will follow that same pattern. The WEF states that there will be an increase of about 78 million new jobs in the year 2030, and that estimate is based on actual research and data from employers. It is not wishful thinking.
The estimate may be correct, but previous innovations inevitably created difficult and painful situations for some people in some industries, in some places, and in some communities. All of this happened even if the economy, the country, and the world prospered. The previous innovations had great impact on the macro scale, but had little to no impact on the micro scale.
Will AI take my job? The most honest answer is: it depends. It depends on the type of job you have, how quickly your employer decides to use AI to do the work for them, how your employer decides to use AI to do the work for them, and if you decide to adapt to the changes by using the skills that are valuable. The answers to all of these questions are in your hands.
Lowest risk of job displacement: Skilled trades (electrical, plumbing, HVAC), mental health counseling, surgical medicine, senior creative direction, frontline physical jobs (construction, agriculture, delivery), advanced negotiation, organizational leadership.
Fastest developing new jobs: AI/ML Engineers, Data Scientists, Cybersecurity Analysts, AI Solutions Architects, Big Data Specialists, AI Prompt Engineers, Healthcare Technology Specialists, Renewable Energy Technicians.
Summery
We are not waiting for AI to change things. Right now, AI influences how companies hire, adjusts the type of talent that is made available, deletes certain roles, adds new roles, and modifies what companies want from their hires.
There is truth to displacement. The opportunities exist, too. The gap between the two is narrowing, and will be determined by the actions of companies, the general workforce, and the policies implemented in the next few years. It is very much in your hands.
