Anthropic, one of the world's leading artificial intelligence companies, says its own workforce has become markedly less junior as AI use has intensified inside the company, according to internal research reported by TechCrunch. Junior staff fell from 21 per cent to 14 per cent of technical employees, and junior appointments dropped from 27 per cent to 9 per cent of technical hires.
Employees reported higher output, but also a substantial new burden: supervising AI-generated work. Meeting time increased by 20 per cent overall and 16 per cent for technical contributors, as human judgement and trust — not raw generation — became the limiting factors in getting work shipped.
In an April 2026 survey cited in the coverage, 27 per cent of employees said the company's Claude AI had made them more productive, while 20 per cent said their job had shifted from doing the work to managing agents — checking outputs, refining prompts and testing systems.
Anthropic cautions, and analysts repeat, that its working practices are unusually AI-intensive and should not be generalized directly to every employer. Even so, the figures challenge the assumption that automation simply removes coordination work — and raise questions about how entry-level talent pipelines will be sustained if junior hiring keeps shrinking.
