Anthropic data: AI to handle most work tasks within a year
New data shows 35% of Claude users expect AI to automate most of their workload by 2027, creating a stark divide between senior and junior roles.
Artificial Intelligence is transitioning from a supportive tool to the primary driver of professional productivity. According to data from Anthropic's June 2026 Economic Index, over one-third of Claude users — based on a sample of 9,700 individuals linked to actual usage data — believe AI will be capable of handling most or nearly all of their work tasks within the next 12 months.
The Experience Paradox
The findings reveal a significant divide across career levels. While senior developers and managers are thriving, leveraging AI to scale their output, entry-level workers face the highest risk of displacement. Anthropic co-founder Jack Clark noted in an interview with
Reason that the company no longer requires junior engineers, as
senior intuition has become far more valuable now that Claude can manage technical execution and experiment scaling.
Eliminating Repetitive Labor
The impact is already evident in complex operational environments. Clark highlighted a case involving the creators of Ozempic, where AI compressed a data-shuffling task from
two months down to one week. While this removal of robotic, repetitive paperwork is seen as a cure for burnout, it raises urgent questions about the future of entry-level employment.
Potential Economic Shockwaves
The global outlook presents a contradiction. AI could drive GDP growth far above historical trends, yet Clark warns this may be accompanied by a
spike in unemployment typical of a recession. This paradox stems from the automation of entry-level roles, which disrupts the traditional professional pipeline. Furthermore, model autonomy is accelerating: Claude's ability to reliably finish autonomous tasks has been doubling every four months, evolving from 4-minute tasks in early 2024 to
12-hour workflows.
Note: AI-Generated Content: This article was created with the support of AI tools and subsequently supervised by the site curator. There may be inaccuracies or missing updates; we recommend verifying original sources before making decisions based on the content.