{
  "id": "F-12",
  "url": "https://evy.so/facts/ai-helps-less-experienced-workers-more/",
  "chapter": "What AI Saves",
  "question": "Does AI help less-experienced workers more than experienced ones?",
  "answer": "Mostly yes, on well-defined tasks: the newest or weakest workers gained about 27 to 46 percent in support, consulting and software studies, against 0 to 17 percent for the strongest. On open-ended business decisions the pattern reversed: in Kenya, strong entrepreneurs gained from an AI adviser while weak ones did nearly 10 percent worse.",
  "range": {
    "low": 27,
    "high": 46,
    "unit": "% gain for the newest or lowest-skilled workers",
    "display": "27–46% gain for the newest or weakest workers; the strongest gain 0 to 17%"
  },
  "confidence": "medium",
  "confidence_reason": "Five independent teams in different settings find bigger gains for weaker or newer workers. Each measures a different outcome, the software gaps are not statistically significant, and a randomized trial on open-ended business decisions finds the reverse.",
  "last_checked": "2026-10-14",
  "how_to_read": "Studies split workers in different ways, by earlier performance or by tenure, and only the support and software studies speak directly to experience. Outcomes differ: issues resolved per hour, graded quality, pull requests and customer ratings. The BCG 43% and 17% compare consultants with their own earlier work and appear in the 2023 working paper; the published 2026 version keeps the finding without printing the numbers. Noy and Zhang and Alibaba show the same narrowing without a comparable percentage, so they are in the table only.",
  "sources": [
    {
      "creator": "Erik Brynjolfsson (Stanford), Danielle Li and Lindsey Raymond (MIT)",
      "title": "Generative AI at Work (Quarterly Journal of Economics)",
      "year": 2025,
      "sample": "5,172 customer support agents at one Fortune 500 software firm",
      "method": "Staggered rollout, compared before and after, split by earlier skill and by tenure. The tenure figure is from the 2023 NBER version.",
      "figure": "Less skilled and less experienced agents +30%; lowest-skilled fifth +36%; agents with under a month of tenure about +46%; no significant change for the most skilled",
      "url": "https://doi.org/10.1093/qje/qjae044",
      "quote": "Less skilled and less experienced workers improve significantly across all productivity measures"
    },
    {
      "creator": "Fabrizio Dell'Acqua, Ethan Mollick, Karim R. Lakhani and colleagues, with Boston Consulting Group",
      "title": "Navigating the Jagged Technological Frontier (HBS Working Paper 24-013)",
      "year": 2023,
      "sample": "758 BCG consultants",
      "method": "Preregistered randomized experiment. Gains compare AI users with their own earlier scores on a baseline task.",
      "figure": "Consultants below the median improved 43%, those above 17%",
      "url": "https://www.hbs.edu/faculty/Pages/item.aspx?num=64700",
      "archive_url": "https://web.archive.org/web/20240102034642/https://www.hbs.edu/ris/Publication%20Files/24-013_d9b45b68-9e74-42d6-a1c6-c72fb70c7282.pdf",
      "quote": "those below the average performance threshold increasing by 43% and those above increasing by 17%"
    },
    {
      "creator": "Kevin Zheyuan Cui, Mert Demirer, Sonia Jaffe, Leon Musolff, Sida Peng and Tobias Salz",
      "title": "The Effects of Generative AI on High-Skilled Work: Evidence from Three Field Experiments with Software Developers (Management Science)",
      "year": 2026,
      "sample": "4,867 developers at Microsoft, Accenture and a Fortune 100 electronics maker",
      "method": "Three randomized company experiments, split by tenure and job level. The splits are not statistically significant.",
      "figure": "Short-tenure developers +27% to 39% completed tasks; long-tenure +8% to 13%",
      "url": "https://doi.org/10.1287/mnsc.2025.00535",
      "quote": "less experienced developers had higher adoption rates and greater productivity gains."
    },
    {
      "creator": "Shakked Noy and Whitney Zhang (MIT)",
      "title": "Experimental evidence on the productivity effects of generative artificial intelligence (Science)",
      "year": 2023,
      "sample": "453 college-educated professionals",
      "method": "Preregistered online randomized experiment on writing tasks",
      "figure": "The gap between stronger and weaker writers halved: low scorers improved in grade and speed, high scorers kept their grade and got faster",
      "url": "https://www.science.org/doi/10.1126/science.adh2586",
      "quote": "ChatGPT compresses the productivity distribution by benefiting low-ability workers more."
    },
    {
      "creator": "Xiao Ni, Yiwei Wang, Tianjun Feng, Lauren Xiaoyuan Lu, Yitong Wang and Congyi Zhou",
      "title": "Generative AI in Action: Field Experimental Evidence from Alibaba's Customer Service Operations",
      "year": 2026,
      "sample": "5,940 new customer service agents, about 2.56 million chats",
      "method": "Four-week randomized trial run by Alibaba, split by earlier customer ratings",
      "figure": "Ratings for the lowest-performing fifth rose 0.81 points on a 5-point scale and fell 0.28 for the top fifth",
      "url": "https://arxiv.org/abs/2603.29888",
      "quote": "Lower-performing agents benefit the most, indicating that gen AI can narrow performance gaps."
    }
  ],
  "related_findings": [
    {
      "question": "When does AI help strong performers more than weak ones?",
      "answer": "On open-ended decisions, where people must judge which advice fits. In a five-month trial with 640 entrepreneurs in Kenya who got a GPT-4 business mentor on WhatsApp, low performers' revenue fell about 9% while high performers' rose about 18%, a gain the authors treat as tentative. They trace the gap to weaker owners acting on generic advice such as price cuts.",
      "sources": [
        {
          "creator": "Nicholas G. Otis, Rowan Clarke, Solène Delecourt, David Holtz and Rembrand Koning",
          "title": "The Uneven Impact of Generative AI on Entrepreneurial Performance: Evidence from a Field Experiment in Kenya (HBS Working Paper 24-042)",
          "year": 2026,
          "sample": "640 entrepreneurs in Kenya, five months",
          "method": "Randomized field experiment: a GPT-4 mentor on WhatsApp or a standard business guide",
          "figure": "No average effect; low performers about 9% lower revenue, high performers about 18% higher (p = 0.07)",
          "url": "https://www.hbs.edu/ris/Publication%20Files/24-042_567db548-5b24-4182-b703-9d5ef1328043.pdf",
          "archive_url": "https://web.archive.org/web/20261009165007/https://www.hbs.edu/ris/Publication%20Files/24-042_567db548-5b24-4182-b703-9d5ef1328043.pdf",
          "quote": "low performers did nearly 10% worse due to the AI assistant"
        }
      ]
    }
  ],
  "numbers_you_may_have_seen": [
    {
      "claim": "AI mostly helps top scientists (Toner-Rodgers, MIT, 2024).",
      "what_the_source_says": "MIT said in May 2025 that it has no confidence in the paper's data and asked for it to be withdrawn."
    },
    {
      "claim": "Below-average BCG consultants improved 43% more than the control group.",
      "what_the_source_says": "The 43% and 17% compare consultants using AI with their own scores on an earlier task without AI."
    },
    {
      "claim": "Novice support agents gained 34% with AI.",
      "what_the_source_says": "That is the 2023 working-paper figure. The peer-reviewed 2025 version says 30% for less skilled and less experienced agents and 36% for the lowest-skilled fifth."
    },
    {
      "claim": "AI levels the playing field for every worker.",
      "what_the_source_says": "It holds in studies of well-defined tasks. In the Kenyan trial on open-ended business decisions, AI widened the gap."
    }
  ],
  "citation": "EVY (2026). Does AI help less-experienced workers more than experienced ones? The AI at Work Fact Book, F-12. https://evy.so/facts/ai-helps-less-experienced-workers-more/ (last checked October 14, 2026). CC BY 4.0.",
  "license": "CC BY 4.0"
}