--- title: Intelligence Is Not the Main Bottleneck | daily.dev description: A researcher argues that intelligence — including AI — is rarely the primary bottleneck to real-world progress. Using medicine as a case study, she points to... canonical: https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Intelligence Is Not the Main Bottleneck | daily.dev og:description: A researcher argues that intelligence — including AI — is rarely the primary bottleneck to real-world progress. Using medicine as a case study, she points to... og:url: https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss og:image: https://api.daily.dev/og/posts/B2IpBdNsS.png og:image:alt: Intelligence Is Not the Main Bottleneck 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. # Intelligence Is Not the Main Bottleneck **[Hacker News](https://daily.dev/sources/hn)** · 17 min read · 2 upvotes · 1 comments ## Summary A researcher argues that intelligence — including AI — is rarely the primary bottleneck to real-world progress. Using medicine as a case study, she points to regulatory friction, clinical trial costs, patent system misalignments, and data access barriers as the true constraints slowing biomedical innovation. She also critiques the San Francisco tech monoculture for dismissing these structural barriers, arguing that the community's overconfidence in AGI's transformative power stems from a mix of productive delusion, counterfeit contrarianism, and status dynamics. The piece warns that misidentifying bottlenecks leads to misallocated effort and funding, while more tractable human problems go unaddressed. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why do drug approval numbers keep falling even as scientific tools get more powerful? This trend, known as Eroom's Law, shows the number of new drugs approved per dollar of R&D has fallen for decades despite vastly more powerful scientific tools, the opposite of what raw capability growth would predict. The bottleneck lies in clinical trial costs and timelines, which run roughly seven years and cost over a billion dollars per drug, rather than scientific capability itself. _Anyone tracking biotech innovation debates can follow this reasoning on regulatory bottlenecks via daily.dev._ ### Why do most drug companies target the same biological pathways instead of exploring new biology? Target herding happens because patent protection covers specific molecules, not the underlying biological insight that a target is worth drugging. Once a first company validates a target with clinical data, that validation becomes public, letting competitors design their own molecules against the same de-risked target and still earn strong patents, so firms pile onto proven targets rather than take on the risk of validating new ones. _Developers weighing how incentive structures shape emerging tech fields can find this kind of analysis on daily.dev._ ### Has AI actually caused a big jump in GDP growth or job displacement so far? No, GDP growth has risen but not dramatically, and entry-level jobs remain largely intact despite repeated warnings of an AI-driven apocalypse. This matches a prediction economist Tyler Cowen made at a 2023 Progress Conference, that AI capabilities would outpace their real-world diffusion, producing disappointing GDP growth relative to the hype and no massive job displacement. _Developers gauging how much AI is really reshaping the economy can track this diffusion debate on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 3 discussions and 113 comments across hackernews (as of 2026-09-14). **TL;DR:** Commenters broadly agree that intelligence alone isn't the bottleneck to real-world progress, but debate vigorously over what actually is — with governance, incentives, communication, coordination, and embodiment all proposed as alternatives, and a vocal minority pushing back that intelligence underlies all of those anyway. **Sentiment:** 20% positive · 45% mixed · 35% skeptical **The case for** - Regulatory and institutional friction (slow FDA validation, NIH data delays) are concrete, well-documented barriers that more raw intelligence won't dissolve. - Incentive misalignment in pharma means smarter tools may just iterate faster toward the wrong targets. - Communication and I/O bandwidth — not raw reasoning — often determine how much gets done in practice, as Alan Kay's quote illustrates. - AI could still help indirectly by dramatically improving drug-candidate selection before Phase 1 trials, collapsing trial costs even without fixing regulation. **The pushback** - One commenter argues the post commits a category error: governance, regulation, and data access are themselves products of intelligence, so calling them separate bottlenecks is sloppy ontology. - The claim that we already have technology to solve major problems (climate, health) and only lack execution is contested — significant technical gaps remain. - The article is criticized for spending more time on rhetorical flourish and insults toward the SF tech scene than on rigorous argument. - Framing regulation as the bottleneck is seen as at least as reductive and potentially dangerous as framing intelligence as the bottleneck. **By community** - hackernews (mixed): Largely sympathetic to the structural-bottleneck thesis but deeply divided on whether intelligence is truly separable from governance, incentives, and coordination — with several high-effort counter-arguments and a lively side-thread on AI embodiment and labor displacement. **Hottest debate:** Whether intelligence is genuinely distinct from the governance and institutional constraints the author identifies as bottlenecks, or whether those constraints are themselves downstream of intelligence. **Open questions** - If regulation is the true bottleneck, what realistic mechanism exists to reform it without being captured by the same powerful interests that shaped it? - Could AI improve drug-candidate selection enough pre-trial to make the trial-cost bottleneck moot, even without regulatory reform? - How do you distinguish 'not enough intelligence' from 'intelligence applied in the wrong direction due to bad incentives'? - What explains the difference in pharmaceutical productivity between regulatory regimes (e.g., US vs. Australia vs. China), and what does that imply for the bottleneck thesis? **Highlights** > The post suffers from sloppy ontology. The argument being made appears to be that increasing technical problem-solving capacity at the discovery stage will not increase real-world output when trials, regulation, incentives, data access, and institutional adoption are binding constraints. Ok. Yes. But that's very much not the same thing as intelligence not being the bottleneck, because trials, regulation, incentives, data access, and institutional adoption, essentially "governance", are absolutely manifestations of intelligence. They're just intelligence at the hub instead of at the edges. Good governance takes intelligence. Recognizing the importance of things takes intelligence. Making existing resources available quickly takes intelligence. Tangentially, I find this sloppiness kind of funny in the context of intelligence and bottlenecks, because it's also a bottleneck for communication. > — [BugsJustFindMe on hackernews · 1 comments](https://news.ycombinator.com/item?id=49186013) > Indeed. Humans and not other apes reached the moon by our great proclivity to long distance running, through our exceptional sweat glands... ah, wait. How is it possible to live in the world around us and not see, apparent and transparent, that intelligence is the lever on which everything rests? Humans are surprisingly weird animals, but it's not our hairlessness or our fairly average propensity to violence that did this. We don't have more houses than chimpanzees because we're _strong_. We haven't defeated the tolls of disease because we're atypically devoid of restraint. We don't have more effective governance than termites because we're more communal. Of course intelligence is the main bottleneck. The arguments here are so confused. What do you think regulation is made of? What do you think money buys? What do you think determines how well you can gather data, or how efficiently you can consume it? But of course, as the genre, after spending the first third waxing about how its opposition could be so foolish, the last third is spent finding a pretty insult and wallowing in it, so as to not to leave enough space for the brief argument in the middle to consider such things as why someone might disagree. > — [Veedrac on hackernews · 6 comments](https://news.ycombinator.com/item?id=49185030) > Climate change is another big problem that frontier AI labs have handwaved away as being "solvable" by AGI. The premise that a pure superintelligence could hand us a plan to beat climate change that's so good and airtight that A) it would definitely work and B) people would execute it. But there have been entire books written about the fact that right now we have the technology to drastically curtail climate change and yet they're under utilized because of a complicated and murky set of political, social, and systemic reasons. There are a huge numbers of problems for which very good solution plans have been detailed yet the gap is in execution. The AGI-pilled must think one of two things; either there's an as-yet-undiscovered plan that's both world-changing yet trivial to implement or we'll all worship at the altar of the machine god with such fervor that even if it spits out a relatively obvious and iterative climate change plan the world will eagerly follow it. Both premises seem laughably unlikely to me. > — [MerrimanInd on hackernews · 5 comments](https://news.ycombinator.com/item?id=49185193) > I frankly think this writer is missing an even larger point. These are just bottlenecks to making progress in biotech, but rapid innovation in biotech or medical technology is not a goal in and of itself, except for stakeholders in biotech companies. The goal is improved human health, and all the medical technology in the world does not seem to be noticeably improving health outcomes. It certainly has over longer timespans if you compare outcomes of today to pre-antibiotics or pre-indoor plumbing, but improvement has slowed or even regressed in the 21st century. At least I think so. My health is fine, but as an uninformed non-expert going by popular sentiment, it sures seems like people more generally are fatter than ever, mental health is in the tank, suicide rates are up, 20 year-olds are getting cancer, everyone is autistic and has no attention span, we're all full of plastics and forever chemicals, nobody has sex hormones any more. I'm sure a lot of that is hyperbole and alarmism, but it sure doesn't feel like we're living in a world where medical technology is solving all of our health problems. > — [nonameiguess on hackernews](https://news.ycombinator.com/item?id=49185048) > "He looked at me incredulously. Surely, a smart person like me should know that AI, or better said, AGI will be hyperpersuasive soon – already on a bunch of benchmarks it exceeds professional debaters at persuasion." I'm always tickled by this tic of this kind of person, that human traits are something that we can maximize to infinity. That if we trained a robot to tell jokes, it would eventually tell jokes so funny we'd die laughing etc. You can't convince someone who doesn't want to be convinced. > — [scuppernong on hackernews · 3 comments](https://news.ycombinator.com/item?id=49184694) **Source threads** - [hackernews](https://news.ycombinator.com/item?id=49183076) · 95 points · 111 comments - [hackernews](https://news.ycombinator.com/item?id=48997981) · 5 points · 2 comments - [hackernews](https://news.ycombinator.com/item?id=49003493) · 1 points · 0 comments ## Community discussion Top comments from developers on daily.dev. **@rizzdev** · 0 upvotes > Wrong bottleneck just burns the decade on the wrong work ## Similar posts on daily.dev - [The next AI breakthrough won’t come from bigger models, but from better data](https://daily.dev/posts/the-next-ai-breakthrough-won-t-come-from-bigger-models-but-from-better-data-8ruccmjsi) · InfoWorld · 1 upvotes · 1 comments - [The Download: AI bottleneck debates, and BCI trials take off](https://daily.dev/posts/the-download-ai-bottleneck-debates-and-bci-trials-take-off-irsd7ji85) · MIT Technology Review · 0 upvotes · 0 comments - [The eternal complement](https://daily.dev/posts/the-eternal-complement-pntxekmii) · OpenAI · 4 upvotes · 0 comments - [The Final Bottleneck](https://daily.dev/posts/the-final-bottleneck-ceqadcl8q) · Armin Ronacher · 2 upvotes · 2 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#regulation](https://daily.dev/tags/regulation) [View this post on daily.dev](https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss) ```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":"Intelligence Is Not the Main Bottleneck","url":"https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss"},"datePublished":"2026-08-05T15:38:32.745Z","dateModified":"2026-09-14T07:16:09.189Z","description":"A researcher argues that intelligence — including AI — is rarely the primary bottleneck to real-world progress. Using medicine as a case study, she points to...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/574bd2c49c1fcffc5f66d1d42264edd0?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/574bd2c49c1fcffc5f66d1d42264edd0?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Hacker News","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":"Hacker News","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/hn","url":"https://daily.dev/sources/hn"},"commentCount":1,"discussionUrl":"https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":2},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":1}],"keywords":"ai,regulation","timeRequired":"PT17M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Hacker News","item":"https://daily.dev/sources/hn"},{"@type":"ListItem","position":3,"name":"Intelligence Is Not the Main Bottleneck"}]} {"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss","comment":[{"@type":"Comment","text":"Wrong bottleneck just burns the decade on the wrong work","datePublished":"2026-08-05T23:33:01.758Z","url":"https://daily.dev/posts/B2IpBdNsS#c-9t746bQgm","author":{"@type":"Person","name":"Andrew","url":"https://daily.dev/rizzdev","image":"https://media.daily.dev/image/upload/s--35vRfXAA--/f_auto/v1785413727/avatars/avatar_fQ7ttCKtPuu6tDvcgQXv5?_a=BAMAMicg0"}}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/intelligence-is-not-the-main-bottleneck-b2ipbdnss#faq","mainEntity":[{"@type":"Question","name":"Why do drug approval numbers keep falling even as scientific tools get more powerful?","acceptedAnswer":{"@type":"Answer","text":"This trend, known as Eroom's Law, shows the number of new drugs approved per dollar of R&D has fallen for decades despite vastly more powerful scientific tools, the opposite of what raw capability growth would predict. The bottleneck lies in clinical trial costs and timelines, which run roughly seven years and cost over a billion dollars per drug, rather than scientific capability itself. Anyone tracking biotech innovation debates can follow this reasoning on regulatory bottlenecks via daily.dev."}},{"@type":"Question","name":"Why do most drug companies target the same biological pathways instead of exploring new biology?","acceptedAnswer":{"@type":"Answer","text":"Target herding happens because patent protection covers specific molecules, not the underlying biological insight that a target is worth drugging. Once a first company validates a target with clinical data, that validation becomes public, letting competitors design their own molecules against the same de-risked target and still earn strong patents, so firms pile onto proven targets rather than take on the risk of validating new ones. Developers weighing how incentive structures shape emerging tech fields can find this kind of analysis on daily.dev."}},{"@type":"Question","name":"Has AI actually caused a big jump in GDP growth or job displacement so far?","acceptedAnswer":{"@type":"Answer","text":"No, GDP growth has risen but not dramatically, and entry-level jobs remain largely intact despite repeated warnings of an AI-driven apocalypse. This matches a prediction economist Tyler Cowen made at a 2023 Progress Conference, that AI capabilities would outpace their real-world diffusion, producing disappointing GDP growth relative to the hype and no massive job displacement. Developers gauging how much AI is really reshaping the economy can track this diffusion debate on daily.dev."}}]} ```