{"slug": "the-principal-agent-vs-agents-problem", "title": "The Principal-Agent vs. Agents Problem", "summary": "A new analysis by venture capitalist Arram Sabeti argues that enterprise AI adoption often fails to boost company productivity because employees capture the gains as leisure time, citing a Goldman Sachs analyst who uses AI to finish a presentation in minutes instead of hours, leaving the firm's output unchanged. Sabeti explains this as a principal-agent problem where the human agent's incentive to hide productivity gains from the enterprise principal can lead to 'shadow AI' and misaligned goals.", "body_md": "The Principal-Agent vs. Agents Problem\nEvery human in the workforce has two crude goals:\n1. Be richer.\n2. Be lazier.\nGet paid 2x more, or keep salary flat? Get paid more!\nGo home at 11:01 PM, or stay up until 5 AM? Go home at 11:01 PM!\nThis isn’t a moral judgment. It’s just…economics.\nWhich brings us to one of the stranger problems with enterprise AI: AI can make every employee dramatically more productive without making the enterprise any more productive.\nImagine a Goldman Sachs analyst. At 11 PM, the boss sends over a presentation with the traditional two-word demand: “please fix.”\nIn the old world, the analyst spends six hours changing fonts, updating charts, reconciling numbers, and moving logos three pixels to the left. The deck is finished at 5 AM.\nIn the new world, the analyst secretly gives it to AI. The deck is finished at 11:01 PM. The analyst goes home and goes to sleep.\nThis is obviously a massive productivity improvement for the analyst.\nWhat changed for Goldman Sachs?\nNothing!\nThe same presentation was delivered. The same analyst is employed. The same salary is paid. The same client is billed. Goldman doesn’t get a bigger fee because its analyst slept six extra hours.\nAI created an enormous economic surplus. The analyst captured 100% of it in the form of leisure.\nThis is the classic principal-agent problem—with a new set of agents.\nThe enterprise is an ethereal “principal.” It wants more revenue, lower costs, faster turnaround, happier customers, etc. But an enterprise can’t actually do anything. It needs human agents—employees—to act on its behalf.\nNow those human agents have AI agents acting on their behalf.\nSo the chain looks something like:\nEnterprise principal -> human agent -> AI agent\nThe enterprise wants more output per dollar. The human wants more dollars per unit of effort. The AI agent generally follows the instructions of the human sitting at the keyboard.\nGuess whose objective function gets optimized first?\nThis is why AI “adoption” inside an enterprise can be wildly misleading. Maybe 90% of employees use AI every day. Maybe every analyst, associate, paralegal, recruiter, consultant, and salesperson has become 5x more productive.\nBut if headcount is the same, output is the same, and revenue is the same, the enterprise has adopted AI technologically—not economically.\nThe employees are richer in time. The principal is not richer in money.\nThis also relates to a point I made recently (\n\n[x.com/arampell/statu…](https://x.com/arampell/status/2086902300854132869)): sometimes the user is not the customer. User = person who actually uses the product. Customer = person who actually pays for the product. Normally, user != customer is a strong negative for product quality. If the user and customer are the same person, the product, sign-up flow, onboarding, etc. all have to be great. If they’re different, the customer can force the user to tolerate an awful product. But AI introduces a different—and more interesting—version of user != customer. The human agent is the user. The enterprise principal is the customer. And their goals are not necessarily aligned. The Goldman analyst might LOVE a product that turns six hours of work into sixty seconds. But the analyst might love it precisely because Goldman doesn’t know how much time it saves. What happens if Goldman finds out that every analyst is secretly producing presentations in sixty seconds? Two logical options: 1. The analyst class can be smaller. 2. The existing analysts can produce 5x more work. Both benefit Goldman. Neither necessarily benefits the analyst. So the analyst has a perfectly rational incentive to use AI—and an equally rational incentive to hide the productivity gain. The best product for the user might be one that the customer can’t see! This is also why banning AI inside enterprises will often just create “shadow AI.” If a tool gives somebody back six hours of sleep, a corporate policy memo is unlikely to stop its use. The tool is effectively part of the employee’s compensation. The real enterprise opportunity, then, isn’t merely to get employees to use AI. They’re going to do that anyway. The opportunity is to get the principal to capture some of the benefit. That might mean selling completed outcomes instead of employee tools. Don’t give the analyst a faster way to make the presentation; make the presentation. It might mean redesigning workflows around the new level of output. If something that took six hours now takes one minute, the deadline shouldn’t remain six hours away forever. It might mean measuring throughput, turnaround time, revenue, resolutions, or other outcomes—rather than counting licenses and declaring victory because “80% of employees used AI this month.” And it probably means sharing some of the gains. If every productivity improvement results in more work, layoffs, or lower compensation, employees will rationally conceal productivity improvements. If employees participate in the upside—more pay, promotion, flexibility, or even permission to go home at 11:01 PM—they have a reason to reveal what AI can actually do. Otherwise, the enterprise will spend billions of dollars buying AI tools that its employees use to work less. AI can make the agent lazier. AI can make the principal richer. The trillion-dollar question is whether it can do both.Sometimes the user is not the customer\nUser = person who actually uses the product\nCustomer = person who actually pays for the product\nIf they are one and the same, then the product, sign-up flow, etc have to be great\nYou know they’re not when you see something like this:", "url": "https://wpnews.pro/news/the-principal-agent-vs-agents-problem", "canonical_source": "https://twitter.com/arampell/status/2087258098427982012", "published_at": "2026-08-12 09:30:13+00:00", "updated_at": "2026-08-12 09:41:51.970671+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-ethics", "ai-policy"], "entities": ["Arram Sabeti", "Goldman Sachs"], "alternates": {"html": "https://wpnews.pro/news/the-principal-agent-vs-agents-problem", "markdown": "https://wpnews.pro/news/the-principal-agent-vs-agents-problem.md", "text": "https://wpnews.pro/news/the-principal-agent-vs-agents-problem.txt", "jsonld": "https://wpnews.pro/news/the-principal-agent-vs-agents-problem.jsonld"}}