Is artificial intelligence replacing jobs? How 17 types of professions are feeling the impact
The fear of automation has haunted the job market for decades. But when OpenAI’s ChatGPT arrived in 2022, the debate left the realm of theory and landed squarely on HR desks, boardroom tables, and of course, LinkedIn feeds full of people worried about their own jobs. 😅
Since then, there has been no shortage of doomsday scenarios. The thought experiment by Citrini Research, called The 2028 Global Intelligence Crisis, imagines unemployment skyrocketing as a result of a destructive cycle fueled by AI. While it is fiction, several real companies laid off workers while investing heavily in artificial intelligence in the four years following ChatGPT’s launch.
Still, recent data from outplacement firm Challenger, Gray and Christmas shows that AI was cited in only 3% of all layoff plans announced since 2023, when the firm began tracking that reason for workforce reductions. In other words, the collective panic grew at a much faster pace than the actual numbers can justify.
But hold on, that does not mean nothing is changing. Because it is, and quite a lot. What is happening is more subtle and, depending on your perspective, might even be more concerning than mass layoffs. The job market is being redesigned from the inside out, role by role, task by task, without most people noticing it in their day-to-day lives.
What the data actually says about AI and employment
Anthropic’s report titled Labor market impacts of AI: A new measure and early evidence, published in March 2026, reveals a significant gap between the perception and the reality of artificial intelligence’s impact. The study compares the theoretical capabilities of large language models with the actual usage of Claude, Anthropic’s AI assistant, across different occupations.
To do this, the researchers created a metric called observed exposure. They identified tasks that could theoretically be automated and then examined how often those tasks appear in Claude’s API traffic for professional purposes, organizing the results by occupation.
The main finding was that there is limited evidence that AI has affected employment so far and that the technology is far from reaching its theoretical capabilities. However, the report found suggestive evidence that hiring of younger workers has slowed in the most exposed occupations. This data point is particularly important because it signals that the effect may be concentrating on early-career professionals.
On another research front, Harvard Business School analyzed job postings from 2019 to 2025, assigning an augmentation score to occupations where generative AI had the potential to complement work because they involved analytical, technical, and creative activities. The study also assigned automation scores to roles most likely to be replaced by generative AI because they involved structured or repetitive work.
The results were revealing: job postings for roles with high automation scores dropped 13% in the years following ChatGPT’s launch, while openings for roles with high augmentation scores grew 20%. In other words, AI is not simply eliminating positions. It is redistributing where the value of human work is concentrated.
Companies that laid off workers and invested in AI at the same time
One of the most telling phenomena in recent years has been watching major companies announce, almost simultaneously, mass layoffs and billion-dollar investments in artificial intelligence. The implicit message is clear: it is not that companies are shrinking, it is that they are swapping out the profile of their workforce.
Amazon laid off nearly 10% of its workforce, roughly 30,000 people, in two rounds of cuts in October 2025 and January 2026. At the same time, it announced plans to invest approximately 200 billion dollars in AI and data centers, plus another 50 billion dollars in OpenAI.
Block, the financial technology company, cut more than 40% of its 10,000 employees in February 2026. CEO Jack Dorsey stated that the company was using intelligence tools to do more with smaller teams.
Oracle let go of thousands of employees globally in March 2026, across multiple departments. The company holds a 500-billion-dollar agreement to support OpenAI’s AI infrastructure.
Atlassian laid off around 1,600 workers, approximately 10% of its headcount, in mid-March 2026. Co-founder and co-CEO Mike Cannon-Brookes wrote on the company’s website that the goal was to self-fund more AI and enterprise sales investments while strengthening the company’s financial profile.
Meta cut roughly 8,000 people in May 2026, redirecting resources, including another 10% of the workforce, toward AI initiatives. This came after 700 layoffs in the Reality Labs unit in March, when CEO Mark Zuckerberg said the company was seeing projects that previously required large teams being completed by a single talented individual.
This pattern repeats across virtually every sector that has adopted automation more aggressively, and it tells a story far more complex than simply more jobs or fewer jobs. 📊
New roles and AI-augmented positions
While part of the debate stays stuck on counting lost positions, an equally important shift is happening on the opportunity side. Job replacement is not the only effect AI has on work. The technology is also automating repetitive tasks and creating roles that simply did not exist a few years ago.
Receive the best innovation content in your email.
All the news, tips, trends, and resources you're looking for, delivered to your inbox.
By subscribing to the newsletter, you agree to receive communications from Método Viral. We are committed to always protecting and respecting your privacy.
- AI model auditors test models and related tools during audits, ensuring they follow company standards.
- Prompt engineers are hired to optimize text inputs for language models, improving generated outputs. They build prompt libraries, establish standards, and correct inconsistencies in responses to refine the models.
- AI ethicists serve as an organization’s moral compass. They guide responsible development, deployment, and oversight of AI systems, ensuring they are safe, fair, and transparent.
- AI architects design the technical structure needed to implement artificial intelligence, including infrastructure and data pipelines. They translate business objectives into functional implementations.
- AI interaction designers shape the decision pathways between human users and artificial intelligence systems.
- Data labelers annotate information to train AI models. Typical tasks include identifying objects in photos, tagging videos, and classifying text.
Beyond these entirely new roles, there is an equally important category: traditional professions that have been supercharged by AI and now require a completely different skill set. AI has empowered some professionals to complete tasks they previously could not and even change the scope of their work.
For example, DevOps engineer Suresh Gangula used TypeScript, Amazon Bedrock, and Claude 4.5 to build a tool that helps his team quickly delete, shut down, and resize services at the company where he works.
On the other hand, adopting AI does not always simplify things. The Guardian reported that Amazon developers use a tool that generates code quickly but end up spending time reviewing frequently flawed AI-generated code instead of writing code from scratch. It is the kind of irony that shows how automation can create new layers of complexity in the workplace. 🤷
The 17 job categories most affected by AI
1. Administrative and office support roles
Generative AI tools can help administrators and assistants with tasks like basic email correspondence, spotting trends in data, finding mutually available meeting times across different time zones, and other summarizing and synthesizing activities. Data entry, typing, and bookkeeping roles have a very high likelihood of being automated.
2. Authors and writers
Tools like ChatGPT and Google Gemini can generate text that reads like it was written by a person. This has serious implications for authors and writers, especially in fields or contexts that demand less nuance, originality, or factual accuracy.
A notable case was the self-published novel Shy Girl, which received positive reviews on the Goodreads website. Publisher Hachette Book Group had planned to release it but pulled back amid allegations that the author wrote it using AI. Some critics argued that fiction can work on multiple levels, and even if the words follow the rhythm of ChatGPT, the premise and overall concept could still make the book a success.
In practice, original or specialized writing tends to become increasingly valued as generic AI-generated text proliferates, obscuring genuinely human perspectives. AI tools can also assist writers with idea development, grammar correction, and high-level research. Over time, readers will likely develop a sharper sense for recognizing the telltale signs of ChatGPT writing.
3. Programming
Anthropic’s study ranks computer programmers as the occupation with the highest level of observed exposure. Hiring of junior developers has declined, and employed developers have added reviewing AI-generated code to their responsibilities. A Harvard study found that when companies adopt generative AI, hiring of junior developers drops dramatically.
Programs like Claude, ChatGPT, and Cursor can write fluent, syntactically correct code faster than most humans. Programmers who mainly produce large volumes of low-quality code are most exposed to replacement. However, those who deliver high-quality products have less to worry about and can use AI to improve their workflows.
4. Customer service
The customer service sector offers many opportunities for automation. AI-powered chatbots provide fast, personalized responses to customer inquiries, theoretically reducing the need for human workers. Other applications include robotic process automation, self-service portals, and sentiment analysis.
Several companies drastically reduced their customer service teams in 2025, including Atlassian, Salesforce, and Sky UK. Klarna laid off hundreds of workers in favor of AI but ran into quality problems and had to rehire staff a year later. That case became an important cautionary tale about the risks of automating roles involving complex human interaction without an adequate backup plan.
5. Drivers and driving assistance
The trucking and automotive industries use AI for driver assistance, accident prevention, route planning, predictive maintenance, and training systems. AI has the potential to create new efficiencies in this area.
In the Harvard Business School study, car and truck drivers scored below average on the automation scale. Industrial truck drivers and taxi drivers were classified as professions less exposed to automation. Even so, it is already possible to hail an autonomous Uber in several American cities. And self-driving trucks are already hauling freight on U.S. highways. Even if trucking jobs could be automated at scale, that automation would likely need to be rolled out gradually, given the logistical and regulatory complexity involved.
6. Legal professions
There is significant evidence that AI will impact legal professions. Most roles in the field, including lawyers and paralegals, scored above average on the HBS automation index.
A 2023 Goldman Sachs study estimated that AI could perform 44% of the tasks that legal assistants in the United States and Europe typically handle. OpenAI’s GPT-4 passed the American bar exam at the 90th percentile. Anthropic also released AI plugins in early 2026 that stirred up the legal industry.
Some experts predict that the legal field will face a dynamic similar to the one affecting programming. Younger legal professionals may struggle to find work, while more experienced ones use AI to automate routine tasks like document review, contract analysis, legal research, and case law searches.
A relevant problem: there have already been numerous cases where AI generated fake legal citations. A database maintained by legal researcher and professor Damien Charlotin identified more than 1,400 court decisions in which courts found that generative AI produced hallucinated content. This is a concrete risk that the profession needs to learn to manage. ⚖️
7. Marketing
Anthropic’s research identified marketing professionals as one of the occupations most exposed to AI replacement. The technology can automate tasks like personalized content creation, customer segmentation, social media management, and data analysis.
Marketing professionals use generative AI tools to create content, personalize emails, and score leads at a speed humans cannot match. AI also assists with search engine optimization tasks, generating optimized meta descriptions and title tags and ensuring a consistent brand voice across all marketing materials.
An interesting example of generative AI in marketing was the #NotJustACadburyAd campaign, which used a digital likeness of Bollywood star Shah Rukh Khan to create thousands of hyper-personalized ads for small local businesses. The campaign featured a microsite that allowed small business owners to create their own version of the ad with the Bollywood star.
Studies indicate that AI is also changing the marketing industry as a whole, with less focus on a unified monoculture and more emphasis on granular segmentation and personalized experiences.
8. Manufacturing
AI on the factory floor is driving significant gains in productivity, quality, and resilience. Still, Cisco’s State of Industrial AI report shows that manufacturers face several barriers to adoption, including cybersecurity concerns, a lack of collaboration between IT and operations teams, and unreliable networks.
Manufacturers adopting AI are focusing on efficiency and throughput applications aligned with short-term cost and productivity goals. This includes process automation, supply chain and logistics automation, and automated quality inspection.
9. Teachers
AI is being used in classrooms to help teachers with resource creation, lesson planning, administration, and grading. It is also being used to teach directly: Alpha School is a network of private K-12 schools in several American cities where students use AI-powered, self-directed learning platforms for core academic instruction.
However, AI also creates new challenges. An immediate concern is that teachers will find it harder to detect plagiarism or other types of cheating in student work. There is also worry that AI may erode students’ capacity for independent and critical thinking.
According to a recent survey of more than 9,000 teachers in the United Kingdom, three-quarters are using AI in their daily work. However, 66% of high school teachers said that students’ critical thinking skills have declined with AI use. That figure deserves special attention from educators and policymakers. 📚
10. Tourism and travel
AI can help travelers discover new destinations and opportunities. AI assistants and chatbots help users book flights, rent vehicles, and find accommodations online, offering a personalized booking experience. AI can also handle flight price forecasting, analyzing historical pricing patterns and telling travelers the best time to buy a ticket.
Tourism companies use AI to analyze the large volume of data their customers generate, such as feedback, reviews, and surveys. Reports predict an impending labor shortage in the sector that AI could theoretically help alleviate.
11. Translators
AI has impacted wages and job availability for interpreters, translators, and product localization professionals, according to Brian Merchant, who publishes a newsletter tracking AI’s effects on jobs. In some cases, companies are hiring translators to edit machine-generated output, a role quite different from original translation.
Merchant noted that AI’s translation capabilities can be limited. Translation requires a nuanced understanding of body language and emotions that AI cannot always provide. This is yet another example of how automation transforms a profession without necessarily eliminating it entirely.
12. Finance
AI is affecting the financial and banking sector broadly. Financial risk specialists, financial services sales agents, credit counselors, accountants, and financial and investment analysts are among the professionals at high risk of automation.
Research from Datarails shows that compensation in most financial job postings has dropped, except for CFOs. The survey also found that 31% of postings mentioned AI or machine learning skills.
Generative AI in finance can be used for financial reporting and summaries, budgeting, expense management, tax preparation and compliance, strategic planning, fraud detection, mergers and acquisitions analysis, and employee training.
AI hallucinations are a significant problem in this sector. Deloitte’s Australian arm was called into question in October 2025 when reports surfaced that a document produced for the Australian government contained AI-generated errors. In finance, where accuracy is paramount, this type of failure can have serious consequences.
13. Engineering
The HBS study found that, among all engineering professions, environmental engineers had the highest potential for automation. Robotics technicians, architects, and cartographers are professions more likely to be augmented by AI. Many engineering roles involve strict compliance and functionality requirements that can be risky to delegate to non-deterministic AI systems.
Generative design is an area where AI is effectively augmenting engineering professions, accelerating the computer-aided design process. It helps with ideation by generating every possible solution to a problem within a specific set of parameters, even when the design is completely novel and a radical departure from anything done before.
14. Human resources
The hype around AI and the fear of losing jobs have created a tough dynamic for HR departments to manage. While dealing with employee anxiety, generative AI also promises to seep into every aspect of HR.
HR professionals already use a variety of AI-powered recruiting tools, along with performance evaluation, analytics, and monitoring tools. A recent Gartner survey estimated that half of all HR activities will be automated by AI by 2030. In the same report, 92% of HR leaders said they had already taken steps to implement AI in the function within the past six months.
HR leaders who focus on upskilling and teaching new competencies, such as AI literacy and intelligent workflow design, tend to better position their teams. It is also important to elevate existing skills like data engineering and change leadership, and to preserve essentially human competencies like emotional intelligence, critical thinking, and data-informed judgment.
15. Retail
Many retail roles scored around or below average on the HBS automation scale. Even so, major retailers are using AI wherever they can.
Amazon plans to build hybrid supercenter-style warehouses powered by robotics and AI. The concept has in-store shopping, pickup, and delivery all operating under one roof. The initiative, called Project Kobe, is in early development. An AI layer will help determine what each store sells, reducing manual planning decisions for managers. Even with AI-driven selection and robotic automation in the warehouses, Amazon anticipates an ongoing need for human workers in the stores.
AI is also changing the way customers shop. Major retailers like Etsy, Target, and Walmart have made their products available for purchase through ChatGPT. Adobe’s Holiday Shopping 2025 report found that generative AI traffic to U.S. retail websites grew nearly 700% year over year. Experts noted that purchases made through agentic tools like ChatGPT and Gemini could make it harder for retailers to collect data, since the AI companies would own that information. 🛒
16. Software quality assurance analysts and testers
This category is on Anthropic’s top 10 most-exposed professions list and also had one of the lowest augmentation scores in the HBS study.
Test Guild, a company that provides learning resources on automated software testing, estimated that more than 80% of development teams use AI in their testing workflows. AI tools can automate the repetitive work of testing, allowing people to focus on issues that require human perspective and judgment. The tools can write tests, visualize applications, and run agentic workflows.
Developers using AI, or anyone with access to these tools, can now generate code faster than testers can validate it. This situation is changing the dynamics of software development. In some cases, development and testing roles are being merged into a single position.
17. Medical transcriptionists
Medical transcriptionists rank high on both the HBS and Anthropic lists as professions with automation potential. AI-powered medical transcription tools are reducing the time spent on data entry and medical documentation.
However, complications exist. Some tools struggle to understand nuances in human speech. Integration with electronic health records has also been a problem. And there is the question of trust: if patients know a doctor is using AI and worry about possible inaccuracies or errors, they may withhold information during the appointment, which creates a real clinical risk.
The silent redesign of the job market
Perhaps the most underestimated aspect of this entire transformation is the speed at which the internal functions of professions are changing, even when the job title stays the same. An accountant in 2026 is still called an accountant, but the tasks they perform day to day are radically different from those of just a few years ago. Software with built-in AI already handles bank reconciliation, automatically classifies entries, identifies inconsistencies, and generates preliminary reports. What is left for the human professional is interpreting the context behind the numbers, talking with clients about strategic planning, and making decisions that involve situational nuance.
This phenomenon is happening across virtually every field at the same time. In healthcare, radiologists work alongside AI systems that analyze images with impressive accuracy, shifting the focus of their work to more complex cases and result validation. In law, AI tools conduct case law research in minutes, something that used to take hours. In design, generative platforms deliver visual drafts in seconds, changing the designer’s role from executor to curator and creative director.
This internal shift within roles is what makes the impact of automation so difficult to measure precisely. Traditional employment metrics do not capture the qualitative transformation of work, only the presence or absence of a position. And that is exactly why looking only at unemployment rates to understand AI’s impact is like trying to understand the ocean by looking only at the surface.
What is happening in the depths, in the way every task, every decision, and every interaction is being mediated by intelligent systems, is where the real revolution is taking place. Quietly, every single day. The question that remains is not whether AI will change your job, but when and how that change will reach your desk. 🌊
