A new benchmark measuring how well artificial intelligence performs actual paid work has delivered a striking result: the best available model can now complete more than four out of every ten real freelance jobs at a professional quality level. That is up from just 2.5 percent a year ago. The leap comes from Anthropic’s Fable 5.2, which scored 41.3 percent on the Remote Labor Index, a test designed to move beyond synthetic tasks and evaluate AI on the kind of projects that freelancers are actually hired to do.
For remote workers, HR teams, and business leaders, the number is more than a technical milestone. It suggests that a meaningful share of freelance work long performed by humans can now be delivered by software quickly and cheaply. At the same time, the remaining gap shows where human judgment and expertise still matter.
From 2.5 percent to 41 percent in a year
The Remote Labor Index measures real freelance jobs rather than simplified benchmarks. A year ago, the strongest AI system completed only 2.5 percent of those jobs at professional quality. Fable 5.2 now reaches 41.3 percent, more than double the next-best systems, including Anthropic’s Opus 5.5 and OpenAI’s GPT 6 Astra. In other words, progress has not been linear; it has accelerated sharply.
That acceleration matters because it changes the ceiling of what organizations can automate. Tasks that once seemed too messy, context-dependent, or subjective for AI—drafting client-facing copy, cleaning a dataset, generating slide decks, writing code for a small business tool—are now within reach for a large minority of real assignments. The benchmark does not claim that AI is ready for all freelance work, but it shows that the boundary has moved far faster than many expected.
What it means for remote and freelance work
The immediate implication is that commodity freelance work has become more exposed. If an AI system can complete 41 percent of real jobs at professional quality, then clients have less reason to pay a human for the straightforward version of those tasks. Freelancers who depend on repetitive writing, basic design, simple data entry, or short programming tasks may face downward price pressure and lower demand.
However, the result also clarifies what remains human. Nearly 59 percent of the jobs were not completed at professional quality. That likely includes work requiring deep context, nuanced client communication, original strategy, accountability, and trust. A client may tolerate an AI draft, but not an AI that misunderstands a brand, mishandles sensitive information, or fails to ask the right questions. Remote workers who position themselves as problem solvers, advisors, and quality owners can still differentiate themselves.
How HR and business leaders should respond
For HR professionals and hiring managers, the benchmark should prompt a review of which roles and contracts are truly human-first. If a job can be completed by AI 41 percent of the time today, that number will likely rise. Leaders should map their workflows and ask where AI can handle first drafts, research, or repetitive production, and where human review, stakeholder management, or creative judgment is irreplaceable.
This also affects hiring criteria. Instead of testing only for the ability to produce a deliverable, organizations may need to test for AI literacy, critical review, client empathy, and the ability to handle exceptions. A freelancer who can manage an AI agent, verify its output, and take responsibility for the final result may become more valuable than one who simply completes tasks manually.
Remote staffing platforms and flexible-work marketplaces will feel this shift too. Platforms like XMF, which connect businesses with remote professionals, may increasingly need to emphasize human judgment, specialized knowledge, and AI-enabled workflows rather than simple task completion. The most resilient professionals will be those who treat AI as a collaborator, not a competitor.
The 41 percent figure is not a prediction that nearly half of freelancers will disappear. It is a signal that the definition of competitive freelance work is changing. The advantage will shift to those who can combine domain expertise with AI tools, deliver quality on complex assignments, and build trust with clients. For businesses, the task is to adopt AI without losing the human judgment that prevents costly mistakes.
Originally published by XMF, inspired by publicly reported industry news.

Likes 0
Save
Copy Link


沪公网安备 31011702008840号
Electronic Business License







