Andrej Karpathy, the former Tesla AI chief and OpenAI cofounder, created an open-source tool that scored every major U.S. occupation on its exposure to current AI capabilities, revealing that 49 million jobs representing roughly $3.7 trillion in annual wages face high AI exposure. The analysis pulled 2024 data from the Bureau of Labor Statistics covering 342 job types and approximately 143 million workers across every sector of the American economy, then ran each occupation through a large language model (LLM), or AI system trained on vast amounts of text, to score it on a scale from zero to ten. Karpathy described the project as a "Saturday morning two hour vibe coded project," referring to his rapid development approach, and emphasized it was designed as a development tool rather than a formal economic forecast. However, the data it generated sparked significant attention and debate about which workers face the greatest risk from AI automation. After the interactive chart drew widespread interest, Karpathy pulled it down, explaining that people had "wildly misinterpreted" his work and were "sensationalizing the visualization tool and putting words in my mouth". Which Jobs Face the Highest AI Exposure? The average exposure score across all occupations was 5.3 out of 10, suggesting that a large share of U.S. jobs already overlap meaningfully with AI capabilities. But the real story lies in the extremes. When isolating occupations scored seven or higher, the numbers become striking: 130 occupations in this high-exposure tier represent 49 million jobs worth approximately $3.7 trillion in annual wages, or more than a third of all jobs tracked by the Bureau of Labor Statistics. The single strongest predictor of high AI exposure is deceptively simple: does the job's output live on a screen? If your work product is fundamentally digital, text-based, code-based, data-focused, or design-focused, then AI can, in principle, replicate or dramatically augment what you do. This explains why the top of the exposure list reads like a roster of knowledge economy jobs, many of which were considered safe, prestigious, and well-compensated just a few years ago. Only one occupation scored the maximum 10 out of 10: medical transcriptionists, with 43,900 jobs at a median pay of $37,550. The rationale is straightforward; this is a purely digital, routine information-processing task where AI speech recognition and large language models have already reached near-human accuracy. The Bureau of Labor Statistics projects employment in this role to decline outright over the 2024 to 2034 period. Thirty occupations scored 9 out of 10, and this is where the sheer scale of exposure becomes apparent: - Customer Service Representatives: 2.8 million jobs at $42,830 median pay, with the core role of answering questions, processing orders, and resolving complaints mapping directly onto what conversational AI already does at scale. - General Office Clerks: 2.6 million jobs at $43,630, where data entry, document formatting, scheduling, and information processing are among the most automatable tasks in any economy. - Software Developers, QA Analysts, and Testers: 1.9 million jobs at a median pay of $131,450, where coding, debugging, and test automation are precisely the tasks where large language models demonstrate the most dramatic productivity gains. - Bookkeeping, Accounting, and Auditing Clerks: 1.6 million jobs at $49,210, already projected to decline over the 2024 to 2034 period. - Financial Clerks: 1.2 million jobs at $48,650, where routine data entry, record updating, and basic financial calculations are textbook cases for AI-assisted workflows. Other occupations scoring 9 out of 10 include market research analysts (942,000 jobs), financial analysts (429,000), paralegals (376,000), graphic designers (266,000), data scientists (246,000), web developers (215,000), writers and authors (135,000), editors (116,000), interpreters and translators (75,300), and computer programmers (121,200 jobs, already in projected decline). What About Professional Jobs Like Lawyers and Accountants? At 8 out of 10, the pattern extends into roles traditionally considered insulated by professional barriers. Secretaries and administrative assistants (3.5 million jobs), accountants and auditors (1.6 million at $81,680), lawyers (865,000 at $151,160), computer systems analysts (521,000), and advertising and marketing managers (434,000 at $159,660) all face significant exposure. Lawyers are a particularly interesting case. At $151,160 median pay and 865,000 jobs, the legal profession represents one of the highest-wage categories in the high-exposure tier. The core of legal work, including research, document review, contract analysis, and regulatory interpretation, is deeply textual and increasingly within reach of large language models. While courtroom presence, client relationships, and professional judgment provide real insulation, the economics of how legal work gets done are already shifting. It's important to note that high exposure does not necessarily mean a job disappears entirely. In many cases, it means AI can handle a growing share of the workflow, pushing humans toward supervision, exception handling, judgment calls, and relationship-heavy tasks. Some occupations may shrink in headcount, while others may grow in total employment even as the work inside them changes dramatically. How to Assess Your Job's AI Vulnerability Understanding where your occupation falls on Karpathy's exposure scale requires examining several key factors: - Digitization Level: Assess whether your work output is primarily digital (text, code, data, designs) or physical. Jobs that can be done entirely from a home office on a computer have inherently higher exposure scores. - Task Routine: Evaluate how much of your work involves routine, repetitive information processing versus judgment calls, relationship management, and exception handling. Routine tasks score higher on exposure. - Employment Projections: Cross-reference your occupation against Bureau of Labor Statistics 2024 to 2034 employment projections. Of the 130 high-exposure occupations, 29 are already projected to decline, including bookkeeping clerks, computer programmers, general office clerks, bill collectors, and insurance underwriters. However, several high-exposure occupations show faster-than-average projected growth, including software developers, data scientists, market research analysts, and operations research analysts. This likely reflects a restructuring dynamic where demand for the output of these roles keeps rising, but AI-augmented workers can produce more of it per person. The Physical Work Advantage: Where AI Struggles The occupations scoring 1 to 2 out of 10 are overwhelmingly physical, hands-on roles: construction laborers, janitors, roofers, electricians, plumbers, carpenters, home health aides, childcare workers, barbers, and bartenders. These 45 lowest-scoring occupations represent about 53 million jobs. The dividing line isn't skill level or education level; it's physicality. A plumber earning $63,000 scores a 2 out of 10, while a software developer earning $131,000 scores a 9 out of 10. The traditional economic hierarchy, where digital knowledge work commands a premium over manual labor, is being directly challenged by the same technology that helped create that premium in the first place. Karpathy emphasized that his project is not a labor market forecast but rather a provocative, structured snapshot of where current AI capabilities overlap most with existing occupations. He acknowledged that "the 'exposure' was scored by an LLM based on how digital the job is. This has no bearing on what actually happens to these occupations, which has to do with demand elasticity and a lot more". The broader context matters here. Tech leaders conducting mass layoffs have cited AI as justification, but critics argue they're using AI as a distraction from corporate bloat and past overhiring, which they say is the real reason for job losses. Anthropic, an AI safety company, released its own findings about labor market impacts of AI earlier this month, finding that computer programmers, customer service representatives, data entry keyers, medical record specialists, and market research analysts were at the highest risk. However, Anthropic also noted that "AI is far from reaching its theoretical capability" and "actual coverage remains a fraction of what's feasible". The scale of potential disruption, if executives are to be believed, could be staggering. ServiceNow CEO Bill McDermott told CNBC that he expects unemployment for new college graduates to reach over 30 percent. For those in the midst of post-secondary education in software development, accounting, or business administration, Karpathy's analysis offers a sobering reality check about the job market they're entering.
