AI’s hidden office role: tackling the paperwork no one else wants
For every headline-grabbing AI breakthrough, there’s a backlog of thankless paperwork—status reports, slide decks, onboarding checklists—that teams avoid like a Monday morning. According to Anthropic’s analysis of 1.2 million Claude Cowork sessions across 600,000 organizations, half of all usage is devoted to just these kinds of tasks. The data suggests that what developers call “the work around the work” is where AI assistants are quietly earning their keep.
The office’s dirty little AI secret
Anthropic’s numbers reveal a split in AI adoption: while software engineers gravitate toward specialized coding tools like Claude Code, the rest of the workforce leans on Cowork for day-to-day office chores. Compiling status updates, drafting internal documentation, and assembling slide presentations dominate the session logs. These are the tasks that don’t move the needle on a product roadmap but still consume hours each week. That imbalance underscores a practical reality: AI’s biggest near-term value may lie in relieving humans of administrative friction rather than chasing cutting-edge innovation.
Why developers still rule the AI stack
The same dataset shows developers prefer focused coding assistants over general-purpose AI chatbots for software tasks. That preference reflects both workflow fit and performance expectations: specialized tools are optimized for syntax, debugging, and repository integration, areas where Cowork’s broader capabilities fall short. For teams, the takeaway is clear: the most effective AI stack is a layered one, pairing domain-specific assistants with general-purpose helpers rather than relying on a single tool for every job.
Why it matters
This isn’t just about saving time—it’s about recalibrating where human effort is most valuable. By offloading routine paperwork to AI, organizations can redirect skilled employees toward strategic thinking and creative problem-solving. The data also hints at a broader shift: AI adoption may accelerate fastest in areas where the pain of manual work is highest, even if those areas lack the glamour of engineering or research. In practical terms, expect more teams to deploy AI not to replace roles but to elevate them, turning “the work around the work” into a background process rather than a bottleneck.
Source: The Decoder. AI-assisted editorial synthesis — TechnoExpress.