AI vs. Freelancers: Who's Winning the Job Battle? Surprising Insights from Online Gig Work

Not long ago, if you needed a speech polished, a document translated, or a logo designed, you would probably have hired a freelancer online. Millions of people did exactly that, turning to platforms like Fiverr and Upwork to pay someone, possibly on the other side of the world, to get the job done.

In 2023, online gig workers were estimated to number between 154 million and 435 million globally, representing as much as 12.5% of the global labour force. However, today, many people are opting for a different approach: they open ChatGPT. Generative AI now acts as a copy editor, translator, illustrator, and research assistant all in one, capable of summarizing reports, writing social media posts, creating presentations, or producing simple logos at virtually no cost.

So, what has happened to the freelancers who used to handle this work?

AI has long been discussed as a threat to jobs and livelihoods. But what's the reality? In this new series, we explore the impact it is already having on different occupations – and how people really feel about their AI assistants.

Some freelancers are struggling, but perhaps surprisingly, others are doing better than ever. Demand and wages have fallen for certain types of online freelance work, with translation, basic copywriting, and simple graphic design being hardest hit. According to one study, demand for freelance writers dropped by up to 30% after the release of generative AI tools, while other research suggests that freelancers highly exposed to AI saw earnings fall by as much as 14%.

Yet, there is also evidence that many freelancers are thriving. Upwork reports that higher-value contracts—those worth more than US$1,000—increased across various disciplines after generative AI arrived. Freelancers using AI-related skills earn around 40% more than comparable freelancers who do not.

How can both of these scenarios be true? The answer becomes clearer when you stop viewing "freelancers" as a single group and instead examine the tasks and skills they perform.

The Commodified vs. Complex Divide

Some freelance work is highly commodified, consisting of narrowly defined, repetitive tasks that can be easily described and compared, such as translating a document, summarizing a report, drafting a press release, or designing a basic logo. These tasks align perfectly with what generative AI excels at—relying on patterns, templates, and predictable instructions. The more a freelancer's work resembles these AI-capable tasks, the more likely it is to face pressure.

In contrast, other freelancers sell a more complex bundle of expertise. For example, a legal translator doesn't just convert words; they understand legal terminology, cultural nuance, and the risks of errors. Similarly, a branding consultant combines design with market research and consumer psychology, and a software developer uses AI to generate code but still needs to grasp the client's business problem to choose the right solution. These workers can leverage AI to automate repetitive parts while focusing on aspects clients value most: expertise, judgment, and trust.

Online Today, In the Office Tomorrow

This dynamic extends far beyond online freelancing platforms. Online labour markets often serve as an early warning system for the broader economy, as they are more transactional and less protected by institutions like long-term contracts or unions. Technological changes manifest more quickly in these open markets, meaning what happens on Fiverr or Upwork today may occur in offices tomorrow.

This trend is already visible in sectors like law firms, consultancy companies, and marketing agencies, where junior employees often handle tasks such as summarizing documents, preparing presentations, drafting reports, or conducting basic research—precisely the areas where AI excels. Recent evidence from the US labour market indicates that younger and less-experienced workers are bearing the brunt of AI-related disruption, while senior workers, who engage in more complex work combining technical knowledge with experience and human interaction, are less affected.

Legal translation is more than converting words from one language to another.

The response should not be to compete with AI on tasks it already performs well. Instead, workers need support in building deeper forms of expertise and combining skills in ways that are harder to automate. This benefits not only workers but also platforms like Fiverr and Upwork, which rely on offering efficient, high-quality work. As routine tasks become automated, these platforms will increasingly depend on workers who provide more than a standardized service.

Platforms can play a role by offering skill-building courses, training resources, and guidance on using AI productively. They could also provide micro-credentials that certify newly acquired expertise, which have been shown to help workers enter online labour markets and increase their earnings.

The challenge is not to prevent AI use but to ensure workers are not trapped in narrow, standardized, and commodified work that can easily be automated. The future of online and onsite work may hinge less on whether we use AI and more on whether our jobs can be reduced to something an AI can easily imitate.