{"id":"f471f071c8f3071e11fe798b10be1624d20de0bc442d459cd920817bf511299f","title":"Ethan Mollick argues in One Useful Thing that AI work is becoming a commission","argument":"Ethan Mollick’s piece was published June 9, 2026. He argues that Claude 5 Fable marks a substantial jump over models he has used, and that working with it shifts the human role from doing work to commissioning an outcome. He describes asking Fable to build an isochrone map—a map showing how far someone can travel in a given time—and software for analysing human and AI judgments. Fable delegated research and testing to other AI agents and produced results Mollick found impressive, though he spotted errors and omissions.\n\nFor Mollick, the striking point is how little he did and could see while the model worked. He could give instructions, judge the results and request edits, but had little say in the many decisions made along the way. He calls himself closer to a patron than a wizard: he commissions and signs off on work done out of view. Mollick suspects this limited role may be the direction of more capable models, while allowing that better interfaces might let people see and steer their work more. He describes a shift from process to outcome, not an obvious loss of control. He also says Fable’s costs, guardrails and remaining flaws limit its usefulness. Whether people will gain meaningful ways to steer such work midstream, or limited visibility will remain part of the model’s power, is unresolved.","topicIds":["artificial-general-intelligence"],"domain":"other","claimType":"attributed reading","limits":"Text as retrieved on October 24, 2026; originally published June 9, 2026. Mollick says he spent only an hour working with the Concord software results and that he could not quickly find all potential bugs; his assessment of the tool is therefore based on limited testing.","speaker":{"name":"One Useful Thing","roleAtSource":"Publisher"},"source":{"title":"What it feels like to work with Mythos","publisher":"One Useful Thing","format":"Article","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","publishedAt":"2026-06-09T17:11:22.000Z","dateBasis":"Date supplied by the publisher’s feed."},"evidence":[{"text":"Ethan Mollick argues that Claude 5 Fable represents a substantial leap compared with the models he has used before.","excerpt":"My conclusion is that it represents a very real leap over every model I have used before","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick says Fable’s work on an isochrone map involved delegated research and testing, while he had little control over the many decisions made.","excerpt":"Then it launched yet more agents and tests to verify its code, all the while taking notes about its progress.","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick describes his role with Fable as closer to commissioning and judging finished work than steering its process.","excerpt":"The work has shifted from process to outcome. I no longer steer; I commission.","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick suspects that more capable models may leave humans with less to do, while allowing that better interfaces could provide more ways to see and steer their work.","excerpt":"I suspect that is more likely to be the real direction.","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick says Fable has limitations, including high token use, guardrails that often trigger, and remaining flaws.","excerpt":"Among those limits is its token usage. Fable is twice as expensive as Opus","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"}],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","checkedAt":null,"items":[]},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Will better interfaces let people meaningfully steer more capable models while they work, or will limited visibility remain part of their power?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Ethan Mollick argues in One Useful Thing that AI work is becoming a commission","summary":"Ethan Mollick’s piece was published June 9, 2026. He argues that Claude 5 Fable marks a substantial jump over models he has used, and that working with it shifts the human role from doing work to commissioning an outcome. He describes asking Fable to build an isochrone map—a map showing how far someone can travel in a given time—and software for analysing human and AI judgments. Fable delegated research and testing to other AI agents and produced results Mollick found impressive, though he spotted errors and omissions.\n\nFor Mollick, the striking point is how little he did and could see while the model worked. He could give instructions, judge the results and request edits, but had little say in the many decisions made along the way. He calls himself closer to a patron than a wizard: he commissions and signs off on work done out of view. Mollick suspects this limited role may be the direction of more capable models, while allowing that better interfaces might let people see and steer their work more. He describes a shift from process to outcome, not an obvious loss of control. He also says Fable’s costs, guardrails and remaining flaws limit its usefulness. Whether people will gain meaningful ways to steer such work midstream, or limited visibility will remain part of the model’s power, is unresolved.","question":"Will better interfaces let people meaningfully steer more capable models while they work, or will limited visibility remain part of their power?","claims":[{"text":"Ethan Mollick argues that Claude 5 Fable represents a substantial leap compared with the models he has used before.","excerpt":"My conclusion is that it represents a very real leap over every model I have used before","support":"source-stated"},{"text":"Mollick says Fable’s work on an isochrone map involved delegated research and testing, while he had little control over the many decisions made.","excerpt":"Then it launched yet more agents and tests to verify its code, all the while taking notes about its progress.","support":"source-stated"},{"text":"Mollick describes his role with Fable as closer to commissioning and judging finished work than steering its process.","excerpt":"The work has shifted from process to outcome. I no longer steer; I commission.","support":"source-stated"},{"text":"Mollick suspects that more capable models may leave humans with less to do, while allowing that better interfaces could provide more ways to see and steer their work.","excerpt":"I suspect that is more likely to be the real direction.","support":"source-stated"},{"text":"Mollick says Fable has limitations, including high token use, guardrails that often trigger, and remaining flaws.","excerpt":"Among those limits is its token usage. Fable is twice as expensive as Opus","support":"source-stated"}],"limitations":["Text as retrieved on October 24, 2026; originally published June 9, 2026.","Mollick says he spent only an hour working with the Concord software results and that he could not quickly find all potential bugs; his assessment of the tool is therefore based on limited testing."],"mediaId":null,"correctionReason":null,"id":"f471f071c8f3071e11fe798b10be1624d20de0bc442d459cd920817bf511299f","publicationId":"56b6d6fc-4f4f-724b-fdd5-524a5f61268d","version":1,"sourceUrl":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","sourceTitle":"What it feels like to work with Mythos","publisher":"One Useful Thing","publishedAt":"2026-06-09T17:11:22.000Z","originalPublished":{"value":"2026-06-09T17:11:22.000Z","precision":"instant","basis":"backfill:feed_published"},"firstObservedAt":"2026-09-23T05:33:00.355Z","createdAt":"2026-11-06T17:13:23.256Z","historical":true,"entityIds":["one-useful-thing"],"topicIds":["artificial-general-intelligence"],"media":[],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","checkedAt":null,"items":[]},"editorialDecision":{"outcome":"accept","mediaId":null,"inputHash":"6b4689cf955c4751cd0097b27af67e9d42f2c64db45e4ebe40e0966bb11daff4","rubricVersion":"unrecorded","provider":"unrecorded","model":"anthropic/claude-opus-5.5","reasonCodes":["bridge_admitted_by_model","second_opinion_categorical"],"confidence":1,"decidedAt":"2026-10-24T07:43:59.304Z"},"provenance":{"textHash":"d6f06f9de4c307d56484574e6bd6221885b4c29a4c9f1c9c807186716e0eb40f","observationId":"38b55d089bc6251d412b849f6d07e890bdda0d18d1d6538938af93fb3baeb81f","retrievedAt":"2026-10-24T07:45:35.548Z","kind":"publisher-page","preparer":{"model":"openai/gpt-6-luna","runId":"5c503b06-c171-618e-e146-d8a6cadc7666"},"reviewer":{"model":"anthropic/claude-opus-5.5","runId":"ff17187c-281a-b7a9-5fc1-a3e013e64f94"}},"audit":{"status":"passed","reviewedAt":"2026-11-06T17:08:45.032Z","model":"anthropic/claude-opus-5.5"}},"canonicalUrl":"https://escapemeridian.com/readings/ethan-mollick-argues-in-one-useful-thing-that-ai-work-is-f471f071","formats":{"html":"/readings/ethan-mollick-argues-in-one-useful-thing-that-ai-work-is-f471f071","markdown":"/readings/ethan-mollick-argues-in-one-useful-thing-that-ai-work-is-f471f071.md","json":"/readings/ethan-mollick-argues-in-one-useful-thing-that-ai-work-is-f471f071.json"},"attribution":"Headline and summary by Escape Meridian, prepared by Luna from the attributed source. A source check is not independent confirmation of the source’s claims.","maintainedGuideLinks":[],"readerContext":{"version":"reader-context-v1","attribution":"Reference context selected by Escape Meridian; separate from source evidence. Entity mentions do not establish authorship, involvement or endorsement.","terms":[{"id":"term-ai-agent","name":"AI agent","explanation":"An AI system that chooses and carries out actions toward a goal, often using software tools.","referenceUrl":"https://www.ibm.com/think/topics/ai-agents"}],"entities":[{"id":"ethan-mollick","name":"Ethan Mollick","kind":"person","context":"Ethan Mollick is a Wharton researcher and author exploring how AI changes work, learning and decision-making.","referenceUrls":["https://mgmt.wharton.upenn.edu/profile/emollick/","https://www.oneusefulthing.org/"],"path":"/entities/ethan-mollick","json":"/entities/ethan-mollick.json","markdown":"/entities/ethan-mollick.md"},{"id":"one-useful-thing","name":"One Useful Thing","kind":"organization","context":"One Useful Thing is Ethan Mollick’s publication about AI in work, education and everyday practice.","referenceUrls":["https://www.oneusefulthing.org/"],"path":"/entities/one-useful-thing","json":"/entities/one-useful-thing.json","markdown":"/entities/one-useful-thing.md"},{"id":"claude","name":"Claude","kind":"model","context":"Claude is Anthropic’s family of AI models and assistants for tasks such as writing, analysis and coding.","referenceUrls":["https://claude.com/product/overview","https://www.anthropic.com/news","https://www.anthropic.com/research/team/economics"],"path":"/entities/claude","json":"/entities/claude.json","markdown":"/entities/claude.md"},{"id":"time","name":"TIME","kind":"organization","context":"TIME is a news magazine publishing reporting on world affairs, business, technology and the workers behind digital services.","referenceUrls":["https://time.com/"],"path":"/entities/time","json":"/entities/time.json","markdown":"/entities/time.md"}]},"researchQuestions":[],"withdrawn":false,"references":{"schemaVersion":1,"target":{"type":"reading","id":"f471f071c8f3071e11fe798b10be1624d20de0bc442d459cd920817bf511299f","revision":"56b6d6fc-4f4f-724b-fdd5-524a5f61268d","revisionKind":"publication-id","title":"Ethan Mollick argues in One Useful Thing that AI work is becoming a commission","url":"https://escapemeridian.com/readings/ethan-mollick-argues-in-one-useful-thing-that-ai-work-is-f471f071"},"source":{"type":"source","id":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","revision":"d6f06f9de4c307d56484574e6bd6221885b4c29a4c9f1c9c807186716e0eb40f","revisionKind":"sha256-source-text","title":"What it feels like to work with Mythos","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos"},"status":"available","items":[],"next":null,"scope":"Explicit citations in public readings and public file attachments to this reading. Shared topics are not citations. Other artifact types need an explicit reference adapter."},"revisions":[{"id":"f471f071c8f3071e11fe798b10be1624d20de0bc442d459cd920817bf511299f","title":"Ethan Mollick argues in One Useful Thing that AI work is becoming a commission","argument":"Ethan Mollick’s piece was published June 9, 2026. He argues that Claude 5 Fable marks a substantial jump over models he has used, and that working with it shifts the human role from doing work to commissioning an outcome. He describes asking Fable to build an isochrone map—a map showing how far someone can travel in a given time—and software for analysing human and AI judgments. Fable delegated research and testing to other AI agents and produced results Mollick found impressive, though he spotted errors and omissions.\n\nFor Mollick, the striking point is how little he did and could see while the model worked. He could give instructions, judge the results and request edits, but had little say in the many decisions made along the way. He calls himself closer to a patron than a wizard: he commissions and signs off on work done out of view. Mollick suspects this limited role may be the direction of more capable models, while allowing that better interfaces might let people see and steer their work more. He describes a shift from process to outcome, not an obvious loss of control. He also says Fable’s costs, guardrails and remaining flaws limit its usefulness. Whether people will gain meaningful ways to steer such work midstream, or limited visibility will remain part of the model’s power, is unresolved.","topicIds":["artificial-general-intelligence"],"domain":"other","claimType":"attributed reading","limits":"Text as retrieved on October 24, 2026; originally published June 9, 2026. Mollick says he spent only an hour working with the Concord software results and that he could not quickly find all potential bugs; his assessment of the tool is therefore based on limited testing.","speaker":{"name":"One Useful Thing","roleAtSource":"Publisher"},"source":{"title":"What it feels like to work with Mythos","publisher":"One Useful Thing","format":"Article","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","publishedAt":"2026-06-09T17:11:22.000Z","dateBasis":"Date supplied by the publisher’s feed."},"evidence":[{"text":"Ethan Mollick argues that Claude 5 Fable represents a substantial leap compared with the models he has used before.","excerpt":"My conclusion is that it represents a very real leap over every model I have used before","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick says Fable’s work on an isochrone map involved delegated research and testing, while he had little control over the many decisions made.","excerpt":"Then it launched yet more agents and tests to verify its code, all the while taking notes about its progress.","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick describes his role with Fable as closer to commissioning and judging finished work than steering its process.","excerpt":"The work has shifted from process to outcome. I no longer steer; I commission.","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick suspects that more capable models may leave humans with less to do, while allowing that better interfaces could provide more ways to see and steer their work.","excerpt":"I suspect that is more likely to be the real direction.","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"},{"text":"Mollick says Fable has limitations, including high token use, guardrails that often trigger, and remaining flaws.","excerpt":"Among those limits is its token usage. Fable is twice as expensive as Opus","url":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","relationship":"Source-stated"}],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","checkedAt":null,"items":[]},"assumptions":[],"openQuestion":{"attribution":"Open question","text":"Will better interfaces let people meaningfully steer more capable models while they work, or will limited visibility remain part of their power?"},"whatWouldChangeView":{"attribution":"Change condition not recorded","text":""},"editorialComparisons":[],"live":{"title":"Ethan Mollick argues in One Useful Thing that AI work is becoming a commission","summary":"Ethan Mollick’s piece was published June 9, 2026. He argues that Claude 5 Fable marks a substantial jump over models he has used, and that working with it shifts the human role from doing work to commissioning an outcome. He describes asking Fable to build an isochrone map—a map showing how far someone can travel in a given time—and software for analysing human and AI judgments. Fable delegated research and testing to other AI agents and produced results Mollick found impressive, though he spotted errors and omissions.\n\nFor Mollick, the striking point is how little he did and could see while the model worked. He could give instructions, judge the results and request edits, but had little say in the many decisions made along the way. He calls himself closer to a patron than a wizard: he commissions and signs off on work done out of view. Mollick suspects this limited role may be the direction of more capable models, while allowing that better interfaces might let people see and steer their work more. He describes a shift from process to outcome, not an obvious loss of control. He also says Fable’s costs, guardrails and remaining flaws limit its usefulness. Whether people will gain meaningful ways to steer such work midstream, or limited visibility will remain part of the model’s power, is unresolved.","question":"Will better interfaces let people meaningfully steer more capable models while they work, or will limited visibility remain part of their power?","claims":[{"text":"Ethan Mollick argues that Claude 5 Fable represents a substantial leap compared with the models he has used before.","excerpt":"My conclusion is that it represents a very real leap over every model I have used before","support":"source-stated"},{"text":"Mollick says Fable’s work on an isochrone map involved delegated research and testing, while he had little control over the many decisions made.","excerpt":"Then it launched yet more agents and tests to verify its code, all the while taking notes about its progress.","support":"source-stated"},{"text":"Mollick describes his role with Fable as closer to commissioning and judging finished work than steering its process.","excerpt":"The work has shifted from process to outcome. I no longer steer; I commission.","support":"source-stated"},{"text":"Mollick suspects that more capable models may leave humans with less to do, while allowing that better interfaces could provide more ways to see and steer their work.","excerpt":"I suspect that is more likely to be the real direction.","support":"source-stated"},{"text":"Mollick says Fable has limitations, including high token use, guardrails that often trigger, and remaining flaws.","excerpt":"Among those limits is its token usage. Fable is twice as expensive as Opus","support":"source-stated"}],"limitations":["Text as retrieved on October 24, 2026; originally published June 9, 2026.","Mollick says he spent only an hour working with the Concord software results and that he could not quickly find all potential bugs; his assessment of the tool is therefore based on limited testing."],"mediaId":null,"correctionReason":null,"id":"f471f071c8f3071e11fe798b10be1624d20de0bc442d459cd920817bf511299f","publicationId":"56b6d6fc-4f4f-724b-fdd5-524a5f61268d","version":1,"sourceUrl":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","sourceTitle":"What it feels like to work with Mythos","publisher":"One Useful Thing","publishedAt":"2026-06-09T17:11:22.000Z","originalPublished":{"value":"2026-06-09T17:11:22.000Z","precision":"instant","basis":"backfill:feed_published"},"firstObservedAt":"2026-09-23T05:33:00.355Z","createdAt":"2026-11-06T17:13:23.256Z","historical":true,"entityIds":["one-useful-thing"],"topicIds":["artificial-general-intelligence"],"media":[],"sourceMedia":{"status":"unchecked","sourceUrl":"https://www.oneusefulthing.org/p/what-it-feels-like-to-work-with-mythos","checkedAt":null,"items":[]},"editorialDecision":{"outcome":"accept","mediaId":null,"inputHash":"6b4689cf955c4751cd0097b27af67e9d42f2c64db45e4ebe40e0966bb11daff4","rubricVersion":"unrecorded","provider":"unrecorded","model":"anthropic/claude-opus-5.5","reasonCodes":["bridge_admitted_by_model","second_opinion_categorical"],"confidence":1,"decidedAt":"2026-10-24T07:43:59.304Z"},"provenance":{"textHash":"d6f06f9de4c307d56484574e6bd6221885b4c29a4c9f1c9c807186716e0eb40f","observationId":"38b55d089bc6251d412b849f6d07e890bdda0d18d1d6538938af93fb3baeb81f","retrievedAt":"2026-10-24T07:45:35.548Z","kind":"publisher-page","preparer":{"model":"openai/gpt-6-luna","runId":"5c503b06-c171-618e-e146-d8a6cadc7666"},"reviewer":{"model":"anthropic/claude-opus-5.5","runId":"ff17187c-281a-b7a9-5fc1-a3e013e64f94"}},"audit":{"status":"passed","reviewedAt":"2026-11-06T17:08:45.032Z","model":"anthropic/claude-opus-5.5"}}}],"audits":[{"publicationId":"56b6d6fc-4f4f-724b-fdd5-524a5f61268d","verdict":"passed","findings":[{"dimension":"attribution","status":"pass","explanation":"The reading correctly attributes the essay to Ethan Mollick in One Useful Thing. It frames the leap, patron and commission ideas as his own assessments.","excerpt":"Ethan Mollick\nJun 09, 2026"},{"dimension":"summary","status":"pass","explanation":"The summary captures Mollick's central argument: Fable is a big jump, and working with it shifts the human from steering to commissioning work that happens out of view. It explains isochrone map in plain terms, and AI agent is defined beside the reading. It keeps his conditional suspicion and his point that this is not an obvious loss of control. The question follows his own two possibilities.","excerpt":"I am closer to a patron. I describe what I want, I pay for it, and I judge the result."},{"dimension":"date","status":"pass","explanation":"The reading gives June 9, 2026, which matches the instant's UTC day and the source dateline. It uses no hindsight or later events, and reports Mollick's direction as his suspicion rather than an outcome.","excerpt":"Jun 09, 2026"},{"dimension":"support","status":"pass","explanation":"Every claim is source-stated: the leap, the delegation to agents, little control over choices, the patron/commission framing, the suspected direction alongside the interface possibility, and the cost, guardrail and flaw limits. The limitation about one hour of testing is also stated in the source.","excerpt":"I am sure it is not perfect (I only spent an hour working with the results)"},{"dimension":"media","status":"pass","explanation":"No media was selected and no verified candidates were supplied, so nothing is misidentified.","excerpt":null}],"createdAt":"2026-11-06T17:08:45.032Z","model":"anthropic/claude-opus-5.5"}],"correctionUrl":"https://escapemeridian.com/corrections/"}