Field Notes ยท Topic

AI & Agentic Systems

Artificial intelligence, generative AI, agentic systems, and the governance, trust, and adoption challenges they create for organizations.

86 notes

  1. AI & Agentic SystemsAI Alignment Has a Stakeholder ProblemAlignment is usually described as alignment with human values. The papers that built the method say it aligned models to a small group of people, and IS research on trust shows that creator, user and system pull in different directions.5 min read
  2. AI & Agentic SystemsIn AI Alignment, the Feedback Channel Is an Information SystemAlignment by human feedback depends on raters, pay schemes, sampling rules and review steps. Agency theory saw this coming, because it treats information as a commodity that has a cost and can be gamed.5 min read
  3. AI & Agentic SystemsAI Alignment Is a Promise the Organization MakesWhen an organization's AI contradicts what the organization says it values, trust and reliance fall. Alignment is also a promise about control and accountability across a supply chain, which brings security into the picture.5 min read
  4. AI & Agentic SystemsAI Alignment Is a Process, and IS Has Studied That Process BeforeIS alignment research spent thirty years moving from alignment as a state to alignment as continual adjustment. AI alignment research is now making the same move, and the older field has lessons to offer.4 min read
  5. AI & Agentic SystemsHuman-AI Collaboration Needs Two Parties Who Know Different ThingsA model and a person who know the same things make a weak team. In the evidence, the gains come from what one side knows and the other does not, and that is an information system design question.4 min read
  6. AI & Agentic SystemsHuman-AI Collaboration Needs a Stopwatch, Not Only a SurveyIn three studies of AI help at work, what people believed about their speed and what was measured did not match, and the gap ran in both directions.5 min read
  7. AI & Agentic SystemsHuman-AI Collaboration Is a Question of Who Hands Work to WhomAlmost every experiment puts the person last, deciding after the AI speaks. Workers, experts and one study that let the model hand out the work suggest that the direction of delegation is a design choice.4 min read
  8. AI & Agentic SystemsHuman-AI Collaboration Is Judged Against the Wrong BarMost human-AI teams beat the person working alone. Far fewer beat the better of the two. The bar we choose decides what we end up calling success.4 min read
  9. AI & Agentic SystemsSuperintelligence Is a Construct Without an Agreed MeasureEvery major definition of superintelligence uses a different yardstick, and expert forecasts move by decades when the question is reworded. For an IS reader that looks like a construct problem before it looks like a risk problem.5 min read
  10. AI & Agentic SystemsControl Over Superintelligence Is Handed Over Before It Is TakenThe careful sources do not describe a machine that seizes power. They describe organizations that delegate decisions one at a time under competitive pressure, and delegation is an IS construct.4 min read
  11. AI & Agentic SystemsIn the Superintelligence Race, More Information Can Mean More DangerA simple model of an AI race finds that teams are safest when they know nothing about each other's capabilities. Openness turns out to have several dimensions, and IS knows something about designing disclosure.4 min read
  12. AI & Agentic SystemsSuperintelligence Risk Is Managed by Organizations, Not Only by ModelsThe most practical part of the catastrophic risk literature is about safety culture, separation of duties and stolen model weights. That is organizational and security ground, and IS already studies it.4 min read
  13. AI & Agentic SystemsWhen the Algorithm Fails, Who Takes the Call?Agency theory explains IT outsourcing and governance through principals, agents, and monitoring. AI rewrites the problem in a way the original framework did not anticipate.6 min read
  14. AI & Agentic SystemsThe Internet of Agents Will Be Built by Platforms That Own the ProtocolThe platform that controls the agent communication protocol will control the Internet of Agents, just as Google owned search and Apple owned mobile.6 min read
  15. AI & Agentic SystemsAI Adoption Follows the S-Curve. Most Organizations Are Still in the Chasm.Rogers showed that innovation adoption follows an S-curve. Enterprise AI is not stalling. It is exactly where diffusion theory predicts.6 min read
  16. AI & Agentic SystemsAI Adoption Spreads Through Social Networks as Much as Tech BudgetsRogers, Granovetter, and Burt explain why the department with the AI champion adopts faster while the department across the building with the same budget does not.7 min read
  17. AI & Agentic SystemsWhy AI Adoption Produces ROI for Some and a Headache for OthersTOE from Tornatzky and Fleischer explains why most AI investments fail while a few transform organizations: the three contexts must align.6 min read
  18. AI & Agentic SystemsAI Agents Create a Principal-Agent Problem at Machine SpeedThe AI alignment problem is principal-agent theory with a new degree of freedom: the agent can act faster than human oversight can keep up.6 min read
  19. AI & Agentic SystemsAI Creates an Ambidexterity Challenge Every Organization Is FailingOrganizations use AI to optimize existing processes while calling it transformation. March (1991) would call this pure exploitation, and the exploration gap is getting dangerous.6 min read
  20. AI & Agentic SystemsAI Chatbots Feel Like the Richest Medium but Behave Like a Search EngineAI chatbots feel like a face to face conversation but function like a search engine with a confidence problem. Media richness theory explains why that mismatch is dangerous.6 min read
  21. AI & Agentic SystemsCopilot Generated More Code and More BugsThirty percent acceptance is adoption, not effective use. Burton-Jones and Grange defined effective use as faithful domain representation, not output.6 min read
  22. AI & Agentic SystemsWhen AI Acts Like a Colleague: Structuration in the Agent EraMultiagent AI systems are not just tools being used. They act, and their outputs become part of the organizational structure that shapes the next cycle.6 min read
  23. AI & Agentic SystemsAI's Energy Problem Is a Sociotechnical Systems ProblemTraining AI models consumes staggering amounts of energy. Trist and Bamforth showed in 1951 why optimizing only the technical subsystem creates a collapse.5 min read
  24. AI & Agentic SystemsAI Ethics Boards and the Pull of Legitimacy Over PerformanceMeyer and Rowan predicted that organizations adopt structures for legitimacy, not performance. AI ethics boards are the most direct example I have seen.6 min read
  25. AI & Agentic SystemsAI Adopted the Way the Garbage Can Model PredictsCohen, March and Olsen showed that solutions search for problems, not the other way around. Most organizations adopted AI exactly that way between 2023 and 2025.6 min read
  26. AI & Agentic SystemsWhy Better Models Alone Won't Solve the Hallucination Trust ProblemHallucination is framed as a retrieval bug to be engineered away. Lee and See would call it a trust calibration failure, and that framing changes the solution entirely.6 min read
  27. AI & Agentic SystemsWhy Doctors Resist Better AI: Identity Matters as Much as AccuracyAI outperforms radiologists at detection but hospitals do not adopt it. Identity theory explains why. The tools that succeed will verify professional identity, not threaten it.6 min read
  28. AI & Agentic SystemsThe Gap Between Fair Metrics and Fair Experience (Amazon's AI Hiring Edition)Kattnig et al. (2024) showed a model can satisfy every fairness metric and still feel deeply unfair. Amazon's AI hiring tool proved it with real damage.6 min read
  29. AI & Agentic SystemsAI Implementation Failures Are Often SociotechnicalBostrom and Heinen said in 1977 that MIS failures are sociotechnical. Half a century later, AI implementation fails the same way.6 min read
  30. AI & Agentic SystemsAI and the Labor Market: What IS Research Can ContributeEconomists study macro displacement. IS researchers can study what actually happens inside organizations when AI changes specific jobs, roles, and power structures.6 min read