Nebuly转发了
"What makes this study significant is that it moves beyond predictions and surveys to show exactly how people are interacting with AI on the job." It's great to learn about Anthropic's use cases but... ain't you curious to understand how people use your own AI? This is exactly what Nebuly is about.
In a groundbreaking study, Anthropic just analyzed millions of real Claude conversations and discovered the gap between AI experimentation and actual workplace integration is WAY wider than expected. Here's what to know: By examining interactions across 20,000 occupational tasks, Anthropic uncovered the first data-driven picture of how AI is really being used at work. What makes this study significant is that it moves beyond predictions and surveys to show exactly how people are interacting with AI on the job. Here's the breakdown: DISTRIBUTION OF CONVERSATIONS ACROSS JOB TYPES: (What kinds of tasks people are asking Claude to help with) ? Computer & Mathematical Tasks: 37.2% ? Arts & Media Tasks: 10.3% ? Education & Library Tasks: 9.3% ? Office & Administrative Tasks: 7.9% ? Life Sciences Tasks: 6.4% ? Business & Financial Tasks: 5.9% ? All other categories: < 5% each SALARY RANGE: Distribution of AI Usage by Salary Range: Most Active Users: Technical roles earning $75,000-$100,000 ? Software developers, data analysts, technical writers ? Heavy use for coding, analysis, and documentation Moderate Adoption: Mid-level professional roles ($50,000-$75,000) ? Marketing managers, business analysts, administrators ? Regular use for content creation and data analysis Limited Engagement: High-wage positions ($200,000+) ? Surgeons, specialized physicians, senior executives ? Minimal use, mostly for administrative tasks Lowest Usage: Entry-level positions ($25,000-$40,000) ? Service workers, retail staff, manual laborers ? Very limited use due to job requirements HOW WE ARE ACTUALLY USING AI: ? Task Enhancement (57.4%): ?- Task Iteration: 31.3% (collaborating with AI to refine work) ?- Learning: 23.3% (using AI to understand concepts) ?- Validation: 2.8% (checking work) ? Task Automation (42.6%): ?- Direct Task Completion: 27.8% (letting AI handle full tasks) ?- Feedback Loop Tasks: 14.8% (iterative problem-solving) Three Key Takeaways: 1. Workers are selectively experimenting rather than fully committing 2. Technical tasks lead adoption, but even there it's not comprehensive 3. People prefer using AI to enhance rather than automate their work ++++++++++++++++++++ When your company is ready, we are ready to upskill your workforce at scale. Our Generative AI for Professionals course is tailored to enterprise and highly effective in driving AI adoption through a unique, proven behavioral transformation. It's pretty awesome. Check out our website or shoot me a DM.