--- title: Fragments: October 28 | daily.dev description: This roundup gathers takes on AI's uneven impact across developers and enterprises. Andrej Karpathy describes a widening perception gap: most people see only a... canonical: https://daily.dev/posts/fragments-october-28-nui1ku3fk twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Fragments: October 28 | daily.dev og:description: This roundup gathers takes on AI's uneven impact across developers and enterprises. Andrej Karpathy describes a widening perception gap: most people see only a... og:url: https://daily.dev/posts/fragments-october-28-nui1ku3fk og:image: https://api.daily.dev/og/posts/nuI1Ku3Fk.png og:image:alt: Fragments: October 28 og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Fragments: October 28 **[Martin Fowler](https://daily.dev/sources/mafo)** · 11 min read · 1 upvotes · 0 comments ## Summary This roundup gathers takes on AI's uneven impact across developers and enterprises. Andrej Karpathy describes a widening perception gap: most people see only a 'derpy chatbot' while a small sliver of professionals feel the curve going vertical. Martin Fowler recaps Foocamp, where coaches of immersion courses said the biggest impact was emotional, not technical, since the tools challenge developer identity. Laurie Voss says today is 1997, comparing the AI boom to the early internet. Annie Vella frames the shift as moving from direct to supervisory control over coding agents. Thoughtworks forecasts its token-sourcing strategy will cut its 2028 token costs from $16-40M down to $6.8-1M. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### why do enterprises struggle to adopt AI agents compared to startups Large enterprises adopt AI agents slowly because their workflows involve many people and years of established process that don't change easily, and security concerns like the 'lethal trifecta' remain. Good AI use also depends on well-structured data, and many organizations don't have their data in decent shape yet, which blocks deeper adoption. _Teams weighing AI rollout risk and data readiness can track this debate on daily.dev._ ### how much could Thoughtworks' token sourcing strategy reduce its AI costs by 2028 Thoughtworks forecasts its token costs will drop from $16 to 40 million down to $6.8 to 1 million by 2028 under its token sourcing strategy. The approach rents frontier models for work needing frontier-level reasoning and runs open-weight models on chosen capacity for IP-sensitive or high-volume workloads, scored across domain differentiation and security and sovereignty axes. _Anyone budgeting LLM spend can follow real cost-reduction strategies like this on daily.dev._ ### what is the difference between direct control and supervisory control in AI-assisted software development Direct control is how developers worked before AI, closing the feedback loop themselves by interpreting requirements, changing code, and acting on results. Supervisory control means a coding agent closes that loop itself, writing code, running tests, and fixing mistakes, while the developer tells it what's wanted, monitors progress, and takes over when needed, per the Endsley and Kiris five-level automation model. _Developers redesigning their workflow around coding agents can explore this framing on daily.dev._ ## Similar posts on daily.dev - [Fragments: July 6](https://daily.dev/posts/fragments-july-6-0q4ip6gyp) · Martin Fowler · 0 upvotes · 0 comments - [Sticker shock has execs rethinking this whole AI thing](https://daily.dev/posts/sticker-shock-has-execs-rethinking-this-whole-ai-thing-ywgulwvsx) · The Register · 0 upvotes · 0 comments - [Fragments: February 25](https://daily.dev/posts/fragments-february-25-vn7svb521) · Martin Fowler · 0 upvotes · 0 comments - [Companies rein in AI spending as running costs mount](https://daily.dev/posts/companies-rein-in-ai-spending-as-running-costs-mount-1dwposyjf) · TechCentral · 0 upvotes · 0 comments - [The token bill comes due: Inside the industry scramble to manage AI’s runaway costs](https://daily.dev/posts/the-token-bill-comes-due-inside-the-industry-scramble-to-manage-ai-s-runaway-costs-glxojg09s) · TechCrunch · 0 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#llm](https://daily.dev/tags/llm), [#ai-coding](https://daily.dev/tags/ai-coding), [#domain-driven-design](https://daily.dev/tags/domain-driven-design) [View this post on daily.dev](https://daily.dev/posts/fragments-october-28-nui1ku3fk) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Fragments: October 28","url":"https://daily.dev/posts/fragments-october-28-nui1ku3fk","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/fragments-october-28-nui1ku3fk"},"datePublished":"2026-10-28T12:59:26.168Z","dateModified":"2026-10-28T15:00:54.784Z","description":"This roundup gathers takes on AI's uneven impact across developers and enterprises. Andrej Karpathy describes a widening perception gap: most people see only a...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e868be0951897ef5168f09b9ecefb685?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/e868be0951897ef5168f09b9ecefb685?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Martin Fowler","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Martin Fowler","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/mafo","url":"https://daily.dev/sources/mafo"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/fragments-october-28-nui1ku3fk","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,llm,ai-coding,domain-driven-design","timeRequired":"PT11M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Martin Fowler","item":"https://daily.dev/sources/mafo"},{"@type":"ListItem","position":3,"name":"Fragments: October 28"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/fragments-october-28-nui1ku3fk#faq","mainEntity":[{"@type":"Question","name":"why do enterprises struggle to adopt AI agents compared to startups","acceptedAnswer":{"@type":"Answer","text":"Large enterprises adopt AI agents slowly because their workflows involve many people and years of established process that don't change easily, and security concerns like the 'lethal trifecta' remain. Good AI use also depends on well-structured data, and many organizations don't have their data in decent shape yet, which blocks deeper adoption. Teams weighing AI rollout risk and data readiness can track this debate on daily.dev."}},{"@type":"Question","name":"how much could Thoughtworks' token sourcing strategy reduce its AI costs by 2028","acceptedAnswer":{"@type":"Answer","text":"Thoughtworks forecasts its token costs will drop from $16 to 40 million down to $6.8 to 1 million by 2028 under its token sourcing strategy. The approach rents frontier models for work needing frontier-level reasoning and runs open-weight models on chosen capacity for IP-sensitive or high-volume workloads, scored across domain differentiation and security and sovereignty axes. Anyone budgeting LLM spend can follow real cost-reduction strategies like this on daily.dev."}},{"@type":"Question","name":"what is the difference between direct control and supervisory control in AI-assisted software development","acceptedAnswer":{"@type":"Answer","text":"Direct control is how developers worked before AI, closing the feedback loop themselves by interpreting requirements, changing code, and acting on results. Supervisory control means a coding agent closes that loop itself, writing code, running tests, and fixing mistakes, while the developer tells it what's wanted, monitors progress, and takes over when needed, per the Endsley and Kiris five-level automation model. Developers redesigning their workflow around coding agents can explore this framing on daily.dev."}}]} ```