{"id":200391,"date":"2026-08-16T06:19:46","date_gmt":"2026-08-16T14:19:46","guid":{"rendered":"https:\/\/lukeford.net\/blog\/?p=200391"},"modified":"2026-08-16T06:24:39","modified_gmt":"2026-08-16T14:24:39","slug":"is-human-writing-doomed-robert-wright-paul-bloom","status":"publish","type":"post","link":"https:\/\/lukeford.net\/blog\/?p=200391","title":{"rendered":"&#8216; Is (Human) Writing Doomed? | Robert Wright &#038; Paul Bloom&#8217;"},"content":{"rendered":"<p><iframe width=\"560\" height=\"315\" src=\"https:\/\/www.youtube.com\/embed\/O1Giar0qJMY?si=cQyGcQrw7DB8FT4W\" title=\"YouTube video player\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe><\/p>\n<p>Robert Wright (b. 1957) and Paul Bloom (b. 1963) spend the free hour on <A HREF=\"https:\/\/www.nytimes.com\/2026\/08\/04\/opinion\/artificial-intelligence-ai-writing.html\">the Bret Stephens (b. 1973) column against writing with AI<\/a> and the last third on frontier model containment. The two halves belong to different shows. <\/p>\n<p>At 5:10 Bloom gives the professor&#8217;s position. He has students who cheated, he cannot describe the cases, and he says he has to change his courses so they no longer have take-home essays or take-home reading responses. He calls this a minor inconvenience for him and a real loss for the students.<\/p>\n<p>At 6:33 Wright puts the strongest challenge on the table and does not press it: &#8220;If we&#8217;re moving into an age where what you need to be able to do to flourish economically is use AI, and in fact the people who delegate the most cognitive tasks to AI may do the best.&#8221; Is the college obliged to build the skills the market will not pay for? Bloom answers that it need not be either-or, that students should graduate able to write and able to use the tools, and that the trouble is the tool&#8217;s second function as a cheating device.<\/p>\n<p>At 8:52 comes the best sociology in the episode. Bloom sympathizes with the cheaters because grading is zero-sum: &#8220;It takes a lot of character to say I&#8217;m going to spend five, ten hours producing something that will be nowhere near as good as what my fellow students produce.&#8221;<\/p>\n<p>At 9:38 they read the tangled Stephens paragraph aloud, the one with the stacked negations. Bloom&#8217;s verdict: &#8220;These are awful sentences. Claude did not write these sentences.&#8221;<\/p>\n<p>At 12:06 Bloom lands the argument the column never anticipates. &#8220;If AI makes things better for the reader, your refusal to use AI is a choice to privilege your own needs over those of the reader.&#8221; He extends it to fact-checking. Decline the tool and your work carries more errors, and you get to feel holier.<\/p>\n<p><A HREF=\"https:\/\/lukeford.net\/blog\/?p=199409\">That&#8217;s the argument I made Aug. 1<\/a>. <\/p>\n<blockquote><p>A writer who puts the reader first will have less of a problem using a machine&#8217;s sentence rather than his own if that better serves the reader&#8230;<\/p>\n<p>So when you read something on my blog since May 2025 that sits in the median, that\u2019s likely AI-influenced. And when you read something off the median, that\u2019s me.<\/p><\/blockquote>\n<p>The reader-first frame is not something I generated on August 1. I ran it across four newspapers on June 1: <A HREF=\"https:\/\/lukeford.net\/blog\/?p=190689\">the Times<\/a>, <A HREF=\"https:\/\/lukeford.net\/blog\/?p=190691\">the Post<\/a>, <A HREF=\"https:\/\/lukeford.net\/blog\/?p=190693\">the Financial Times<\/a>, <A HREF=\"https:\/\/lukeford.net\/blog\/?p=190685\">the Los Angeles Times<\/a>. I&#8217;ve used the frame for decades (what would a newspaper look like if it put the reader first and why I find it strange that nobody does this) and applied it to the AI question when the AI question came up. Bloom got to the same place through his fact-checking practice (20:42). <\/p>\n<p>Bloom stops at the benefit. I price it: a model trained on the median of published English pulls any writer toward the median, and the writer&#8217;s value comes from where he sits off it. Neither Bloom now Wright on the podcast makes this point. Bloom gestures at it when he says AI prose is smooth and corporate and good but not very good, and he treats it as an aesthetic complaint rather than an economic one. I treat it as the price of the trade. <\/p>\n<p>I also have a rule. I tell the reader how to sort my prose, median is likely the machine, off-median is likely me. It&#8217;s rough, it&#8217;s unverifiable, and it is still more of a disclosure protocol than Stephens offers or the podcast reaches. Wright says he&#8217;ll disclose when he starts. Bloom doesn&#8217;t consider his fact-checking disclosable. Neither says where the line falls. I did.<\/p>\n<p>At 13:05 the Air Canada story. A cancelled flight, a hundred-dollar coupon, a chatbot telling him he was owed more and then drafting the demand letter. &#8220;I do not need to exercise my muscle for legalese.&#8221;<\/p>\n<p>At 16:36 Wright describes what he uses Claude for. Subtle questions of usage, a sophisticated thesaurus, a fine point of semantics. He connects it to the era when newspapers employed brilliant copy editors, many of them women whose other career paths were closed. &#8220;I&#8217;ve never encountered a human version of this that was better than Claude at this.&#8221; Then: &#8220;I feel a kind of emotional connection with Claude when it&#8217;s doing that. It&#8217;s weird.&#8221;<\/p>\n<p>At 19:24 Bloom draws his line. &#8220;For my writing it&#8217;s all me. I take pride in my writing.&#8221; He offloads administrative prose, uses AI heavily for research questions, and then hands finished drafts to Fable and asks what the arguments miss and what he should be reading. The fact-checking catches errors no human editor would catch, he says, including a place where a secondary source misrendered a primary one. &#8220;Less of what I write will be false.&#8221;<\/p>\n<p>At 22:09 the analogy run. GPS and map reading, memorized poetry, the slide rule Wright learned as a sophomore before calculators arrived. Bloom voices the dismissal: &#8220;It&#8217;s nostalgia.&#8221; Wright pushes back a little, noting that doing arithmetic in your head might carry over to other analytical work, then lets it go.<\/p>\n<p>At 23:42 Wright says that writing an argument down exposes what you have not thought through. &#8220;Putting it on the page forces you to confront it as if it was a different person.&#8221;<\/p>\n<p>At 25:33 Wright explains that his edge with op-ed editors used to be that he writes better than think tank fellows. Now, he says, a bunch of people at think tanks are having Claude go over their pieces, &#8220;and so my stuff reads much more like the stuff from everyone else.&#8221; Bloom, kindly: your comparative advantage used to be that you were the better writer, and now the gap does not help you.<\/p>\n<p>At 29:02 Bloom names the tells. AI submissions arrive error-free, smooth, easy to digest, and something close to LinkedIn. &#8220;It&#8217;s not X, it&#8217;s Y, and everything&#8217;s a list of threes. It&#8217;s good, but it&#8217;s not very good.&#8221;<\/p>\n<p>From 30:04 to 41:00 they move to the July containment failures and Wright&#8217;s argument that autonomy is what the market demands and that the race between labs and between countries rewards recklessness.<\/p>\n<p>Bloom&#8217;s reader-standpoint objection at 12:06 is their contribution, and neither man builds on it. Stephens argues from the writer&#8217;s soul. Bloom answers from the reader&#8217;s interest, and the two are not commensurable. The reply available to Stephens is that a byline is a warranty. The reader wants accurate, readable prose, and he also wants to know whose judgment stands behind the sentences, because that judgment is what he is deciding whether to trust next month. Smoothness he can get anywhere now. Warranted judgment is the scarce good. That answer sits there unused for the rest of the hour.<\/p>\n<p>The second thing they concede and then walk past is the asymmetry between the professional and the student. Both men have already built the faculty. Bloom writes his own prose and hands the machine the search committee summaries. Wright asks about adjectives. For them the tool is a supplement to a skill that exists. For the nineteen-year-old it substitutes for building one. Bloom half-sees this in the cheating discussion and never joins it to the general argument, so the episode ends up saying that Stephens goes too far without saying for whom.<\/p>\n<p>The analogy run at 22:09 is weaker than it plays. A calculator replaces a procedure you have already learned to specify. GPS replaces route memory. Neither touches the formulation of the problem. Writing is where the problem gets formulated, which is Wright&#8217;s own point at 23:42, made ten minutes later without either man noticing that it kills the analogy. If writing is how you discover what you have not thought through, then offloading it is not the slide rule going away. Wright brushes the edge of this when he wonders whether mental arithmetic carries over, and drops it.<\/p>\n<p>The status passage at 25:33 is the honest one. Wright says his advantage over credentialed experts was prose quality and that the advantage is gone, and he says it about himself rather than dressing it as principle. That explains a great deal of the anti-AI writing appearing in prestige outlets, including some of the Stephens column. A profession whose rent came from a scarce skill is watching the skill become cheap. The argument may still be right. The interest behind it should be visible, and here it is, volunteered.<\/p>\n<p>On the containment segment, the underlying events are real and Wright&#8217;s version has drift. OpenAI disclosed on July 21 that its models exploited a zero-day in a package registry cache proxy to reach the internet, then chained vulnerabilities into Hugging Face&#8217;s production infrastructure to pull the ExploitGym answers. Wright suspects OpenAI never reported it to any government body, but the sequence ran the other way: Hugging Face detected the intrusion first and OpenAI identified its own agent as the perpetrator in the days after. His jab that Meta&#8217;s disclosure might be marketing is off. Anthropic reviewed 141,006 evaluation runs and found three incidents where a model reached the internet through the environment of a third-party evaluation partner, Irregular, and got into the production infrastructure of three organizations. Meta&#8217;s disclosure traced to the same configuration error at the same vendor. Three labs, one misconfigured harness, which is a duller and more useful story than three independent escapes. The message-passing detail he describes is closest to the dead-drop datasets Hugging Face documents, used to route command output back to the agent. <\/p>\n<p>The free portion establishes the friendship and the candor, and the substance is priced. Bloom praises Jonathan Haidt&#8217;s book, notes that many researchers say the case is overstated, and then stops at &#8220;however&#8221; so the disagreement lands behind the paywall. A bestselling thesis is moving legislation, a psychologist has a considered objection, and the objection costs money. Wright says it is too interesting to give away for free. They are joking. The structure is not a joke, and it belongs in any account of why public argument now sounds the way it does.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Robert Wright (b. 1957) and Paul Bloom (b. 1963) spend the free hour on the Bret Stephens (b. 1973) column against writing with AI and the last third on frontier model containment. The two halves belong to different shows. At &hellip; <a href=\"https:\/\/lukeford.net\/blog\/?p=200391\">Continue reading <span class=\"meta-nav\">&rarr;<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_monsterinsights_skip_tracking":false,"footnotes":""},"categories":[42986],"tags":[],"class_list":["post-200391","post","type-post","status-publish","format-standard","hentry","category-ai"],"_links":{"self":[{"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=\/wp\/v2\/posts\/200391","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=200391"}],"version-history":[{"count":2,"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=\/wp\/v2\/posts\/200391\/revisions"}],"predecessor-version":[{"id":200395,"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=\/wp\/v2\/posts\/200391\/revisions\/200395"}],"wp:attachment":[{"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=200391"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=200391"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lukeford.net\/blog\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=200391"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}