Correction and context (3 October 2026): The AI skills figure below is historical. Upwork published the 109% figure on 4 February 2026, using U.S.-originating marketplace earnings from 2025. It is not a new October hiring release or a measure of all global job postings. This article has been revised to state the date, method and limits clearly.
AI skills are specific abilities used to apply artificial intelligence to paid work. AI skills explicitly named in work on Upwork grew 109% year over year in the platform’s 2025 earnings data, the company reported in February. The useful business question now is what that figure actually measured: paid work that applied AI inside established professions, not a count of every new job or proof that human expertise has lost value.
Key takeaways
- The 109% growth came from freelancer earnings on contracted jobs with U.S.-originating demand during 2025, compared with 2024.
- Upwork reported especially rapid growth in AI video generation and editing, integration, data annotation and chatbot development; these are growth rates, not rankings by total hiring volume.
- The platform’s survey found many business leaders value creativity and innovation, but it did not establish a universal wage premium for every AI-using freelancer.
- Indian professionals can use the report as a signal of global client interest, while recognizing that its sample is not an India hiring survey.
AI skills growth: what the 109% number means
Upwork’s In-Demand Skills 2026 report grouped six skills whose names explicitly reference AI and compared marketplace earnings for those skills in 2025 with earnings in 2024. Upwork calls the resulting increase 109%. Its comparison group of other in-demand skills rose 23%. These figures describe activity on one work marketplace, within a defined method, rather than the entire labor market.
That distinction matters because a larger percentage can start from a smaller base. A skill growing quickly may still account for less total work than a mature occupation with slower growth. Upwork itself says the fastest-growing skills are not necessarily the skills that produced the most hiring volume. Reading 109% as a broad count of newly created jobs would overstate what the research can support.
The company says the underlying sample covered freelancer earnings across six work categories from 1 January through 31 December 2025. Demand had to originate in the United States, and only contracted work was analyzed. To appear among the top skills, a category had to clear $100,000 in aggregate earnings during the period. The earnings measure is useful because it reflects paid activity. It also excludes unawarded job posts, work outside the platform and clients in other markets.
Upwork released the report on 4 February 2026. Fast Company interviewed Upwork research manager Gabby Burlacu later that month and emphasized a practical point: clients sought people who could apply AI within an existing discipline, rather than generalists who only knew a tool. Entrepreneur separately reported the gap between AI-linked skills and other categories. Quartz examined the operational skills behind the headline.
Which AI skills were growing fastest?
Upwork put AI video generation and editing at 329% year-over-year growth, followed by AI integration at 178% and AI data annotation and labeling at 154%. AI image generation and editing grew 95%, while AI chatbot development rose 71%, according to the company’s report. The numbers are vendor-published platform results, not independent measurements of every employer’s hiring.
The mix is revealing. Video work combines software with editing, pacing and brand judgment. Integration means connecting AI functions to an existing application or business process. Annotation supplies labeled data that a model or evaluation process can use. Chatbot development includes the design, testing and deployment of conversational systems. In each case, the AI component is attached to work a client can specify and pay for.
Entrepreneur’s February coverage described these as practical workflow skills. Quartz also highlighted integration and data preparation. These reports corroborate the categories and point to the same primary Upwork data; they should not be mistaken for three separate independent datasets. The underlying 109% estimate remains Upwork’s own measure.
For a buyer, this distinction changes the hiring brief. Asking for an ‘AI expert’ is too broad to establish what will be delivered. A brief for video editing can specify footage quality, aspect ratios, turnaround and review steps. A brief for integration can describe the systems to connect, permissions, tests and fallback process. A brief for annotation can set labeling guidelines and acceptance thresholds. Those are practical examples of how to define work, rather than claims that Upwork measured the success of a particular project design.
| Skill | Year-over-year change | How to read it |
|---|---|---|
| AI video generation and editing | +329% | Fast growth from its own base |
| AI integration | +178% | Demand for implementation work |
| Data annotation and labeling | +154% | Data preparation remained important |
| AI chatbot development | +71% | Growth, not total job share |
Why human expertise remains part of the story
Upwork also found established skills, including full-stack development, general virtual assistance, data analytics and graphic design, remained prominent on its platform. The company interprets that stability as evidence that AI is being added to familiar roles, rather than simply replacing them. This is a narrower claim than saying no jobs are at risk. Its data does not test every occupation or measure employment across an economy.
A separate survey within the report asked 349 business leaders in October 2025 about talent and skills. Nearly half said they would pay a premium for creativity or innovation, according to Upwork. That is a statement of surveyed intent. It is not evidence for the previously published claim on this page that every AI-using specialist earns 38% more per hour. The earlier page also attributed a 64% automation-bottleneck statistic to Upwork without a traceable basis in this report; that assertion has been removed.
Fast Company’s interview with Burlacu offers a more careful interpretation: expertise in a particular field gives an AI-enabled worker context to judge the output. The report’s evidence supports a demand signal for applied skills, while the interview explains why clients might value the person who integrates the tool into a workflow. Neither proves that a particular worker will earn more simply by adding ‘AI’ to a profile.
A useful way to evaluate human contribution is to ask who defines the objective, who checks the result and who is accountable when the output is wrong. In software work, a developer may test the behavior of generated code. In design, an editor may reject an image that misstates a product. In data work, a reviewer may catch labeling inconsistencies. Those are illustrative responsibilities, not outcomes directly quantified by the 109% figure.
What this means for Indian freelancers and businesses
The report is relevant to Indian professionals who serve clients abroad, but its sample cannot directly describe hiring within India. Upwork states that demand in the analysis originated in the United States. An Indian freelancer could be part of the supply on the platform, yet the report does not publish a separate India demand breakdown for the 109% figure. Translating it into an India-wide wage or employment forecast would therefore be unjustified.
The signal that does travel is the type of work clients paid for: applying AI to a defined production process. For Indian service firms and independent workers, a portfolio that shows a completed integration, a reviewed video workflow or a documented evaluation method may explain capability more clearly than a generic list of tools. That is our practical inference from the categories, not a statistic reported by Upwork.
India has separate evidence on domestic AI skills. Lapaas Voice previously covered Naukri’s report on AI mentions in Indian technology listings and an India agentic-AI skilling programme. Those sources address the local market more directly. They should be read alongside Upwork’s U.S.-originating marketplace data, not blended into a single global percentage.
For businesses, the lesson is to specify a measurable deliverable and its safeguards before hiring. A client deploying a chatbot can ask for a test set, escalation rules and documentation. A team commissioning video can set criteria for factual accuracy, brand use and review. A firm connecting AI to customer data can require clear permissions and a rollback plan. Such requirements make a contractor’s domain judgment visible and reduce the risk of buying an impressive demo that does not survive routine operations.
For workers, a credible AI skills portfolio should identify the problem, the baseline process, the human decisions made, and the output that the client approved. This is particularly useful when results are confidential: a worker can describe the method without exposing private data. The report does not guarantee that any one portfolio style raises earnings; it does show where marketplace spending grew in the measured period.
The limits of the 2026 Upwork report
Upwork’s method uses earnings on contracted jobs, so it gives a stronger spending signal than a count of advertisements. It also has boundaries. It is platform-specific, limited to U.S.-originating demand in the stated analysis, and focused on selected skills. It cannot by itself measure all AI hiring, determine whether employment rose economy-wide, or show what happened after December 2025.
The data also separates growth rates from size. A category rising 329% may have started small. A large mature category can add more actual spending with a much lower percentage increase. Upwork did not publish a simple ranking of all freelance occupations by earnings in the cited headline. Readers should resist turning the fastest-growing list into a list of the largest labor markets.
Finally, survey statements and marketplace transactions answer different questions. The survey records what a group of business leaders said they value; the earnings data records completed work. A statement that leaders would pay extra for creativity is not proof that a particular specialist got a higher rate. Likewise, the 109% figure does not show whether AI caused the growth or whether other demand changes contributed. The article therefore attributes the figures to Upwork and avoids presenting a hiring forecast as a fact.
Frequently asked questions
Did Upwork release the 109% figure in October 2026?
No. Upwork published its In-Demand Skills 2026 report on 4 February 2026. The data compare earnings from 2025 and 2024. This October article explains the older result and its limits; it is not reporting a new release.
Does 109% mean the number of AI job posts doubled?
No. Upwork describes the comparison as growth in freelancer earnings for a group of skills explicitly referencing AI, drawn from contracted jobs with U.S.-originating demand. It should not be rewritten as a count of all job posts or global vacancies.
Which AI skills grew fastest in the report?
Upwork reported AI video generation and editing at 329%, AI integration at 178%, and AI data annotation and labeling at 154% year over year. These are growth rates within its platform, not a statement that those jobs are the largest by volume.
Does the report prove AI users earn 38% more?
No. The February In-Demand Skills report does not establish a universal 38% hourly premium for AI-using workers. The earlier version of this article made that claim without a verified supporting source, and it has been removed.
Sources: Upwork’s research and methodology; independent coverage by Fast Company, Entrepreneur, and Quartz.
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
