AI’s Limits Exposed: New Study Finds Machines Struggle With Real Remote Work
Research shows AI agents fail most remote tasks, with top performer automating just 2.5% of freelance work.
The study, called the Remote Labor Index (RLI), represents one of the most detailed attempts so far to measure AI’s performance on practical digital work.
It focuses on tasks that mirror real online freelancing jobs rather than theoretical tests or benchmark problems.
- Researchers collected 240 completed projects from professional freelancers working through platforms such as Upwork.
- Each project included the original brief, all input materials, and the final deliverable that a client had accepted.
- These projects came from 23 categories of work, including product design, animation, architecture, game development, and data analysis.
- Together they covered more than 6,000 hours of paid labor valued at about $140,000.
Six advanced AI agents were then tested on the same projects.
- Manus
- Grok 4
- Sonnet 4.5
- GPT-5
- ChatGPT agent
- Gemini 2.5 Pro
Author summary: AI agents struggle with real remote work tasks.
More News
- Deepti Sharma stars with bat and ball to help India to World Cup glory
- Why Dentsu’s next buyer can’t look like Havas or the industry it’s leaving behind | Advertising | Campaign Asia
- Opinion: Personal consumerism lets us reclaim joy in corporate slog, so go buy that Labubu - Daily Bruin
- Kangana Ranaut’s Farmer Post Row Becomes A Study In Free Speech And Public Sensitivity
- Jay Glazer says Eagles won't trade A.J. Brown, are interested in Jaelan Phillips | NorthPennNow
- Attorney General orders detention of Acting Education Minister Ali Al-Abed over schoolbook case
- Barcelona v Elche live: La Liga text updates, stats, goals
- ‘My father died and I opened a scrapyard’—the entrepreneur who turned family grief into a million-dollar fortune