{"id":4641,"date":"2026-04-07T06:38:33","date_gmt":"2026-04-07T06:38:33","guid":{"rendered":"https:\/\/frontlinenewsng.org\/?p=4641"},"modified":"2026-04-07T06:38:33","modified_gmt":"2026-04-07T06:38:33","slug":"gary-the-human-in-the-loop","status":"publish","type":"post","link":"https:\/\/frontlinenewsng.org\/?p=4641","title":{"rendered":"Gary: The Human in the Loop"},"content":{"rendered":"

If you\u2019ve been paying attention to LLM-based coding tools in the past few months, you\u2019ll have seen a seismic shift in how they\u2019re being used. Even 12 months ago, they were little more than glorified auto-complete tools: useful for quickly repeating patterns, but terrible at producing well structured, thoughtful, maintainable code. More recently, however, there seems to have been a new equilibrium reached, where an experienced engineer can guide these tools to consistently produce high quality code. Small course adjustments seem to have an outsized effect, resulting in the \u201cHuman in the Loop\u201d paradigm that\u2019s become so popular.<\/p>\n

Why It Works<\/h2>\n

\u201cCode is Poetry\u201d has been my approach to writing code for as long as I can remember. Software is a form of expression, and the way you create that expression is through code. So, to make beautiful software, you need to write beautiful code. But, what happens when you don\u2019t need to write code to create the software?<\/p>\n

Suddenly, the code becomes entirely about outcomes. It needs to be correct, functional, and maintainable, but it doesn\u2019t need to be seen as a form of expression itself. Instead, the creative decisions move further up the stack, to the architectural level. You can write beautiful software by writing thoughtful specifications, instead.<\/p>\n

That\u2019s not to say that technical abilities are suddenly obsolete. You still need to know what\u2019s possible and realistic to be able to tell the LLM what to build, and to redirect it when it goes in a different direction. You need to be able to read and comprehend the code, you just don\u2019t need to memorise every function signature.<\/p>\n

The Temptation<\/h2>\n

So, if an LLM can write code for me, what else can it do? Marketing copy? Emails? Opinion blog posts? I could ask Claude to write 10 paragraphs on the \u201cThe Human in the Loop\u201d, but would you have even read this far if you thought this post was LLM generated? Of course not! I can promise you that every word of this post (and every other post on my blog) was written by me.<\/p>\n

Respect for the Reader<\/h2>\n

If I want you to read this post, and seriously consider the arguments I\u2019m making, the least I can do is write it myself. It goes beyond that, however. LLMs can write functional code, but they can\u2019t write beautiful software. When the text is<\/em> the creative act, there\u2019s no way for the LLM to write the text without compromising your creativity. If you\u2019re the Human in the Loop for a blog post, you\u2019re not injecting your voice, your perspective, or your personality into the post: you\u2019re rubber stamping whatever feels good enough, and that\u2019s a very low bar to clear.<\/p>\n

\n

\u201cGood enough\u201d isn\u2019t actually good enough. <\/p>\n<\/blockquote>\n

A measure of the complexity of a written piece of text is called \u201cperplexity\u201d. It measures the randomness of how the text flows, and it\u2019s probably the thing you\u2019re noticing when you know you\u2019re reading LLM-generated text, but you can\u2019t quite articulate why. It\u2019s an uncanny valley thing: it looks like writing, it reads like writing, it might even flow like writing, but the vibes are off.<\/p>\n

The good news is, you\u2019re not going insane, recent research shows that there is a measurable difference between human written text, and LLM generated text<\/a>. LLM generated text is inherently less random, which makes sense when you remember that LLMs are, at their core, giant statistical models that are really good at figuring out \u201cwhat\u2019s the most likely bit of text to come next\u201d.<\/p>\n

\"\"<\/figure>\n

The LLM as the Assistant<\/h2>\n

That\u2019s not to say that LLMs are completely useless when it comes to writing, but we need to use them the right way. While they shouldn\u2019t be generating text, they can absolutely be used to help you write. Over the last month or so, I\u2019ve been working on Claudaborative Editing<\/a>, an experiment to see exactly how much they can help with the writing process. I\u2019ve been building it directly into the WordPress editor, allowing me to plan, write, review, and publish this post from the one place. An LLM assisted, but every word of it was written by me alone<\/em>. My goal isn\u2019t to replace the author, or to make it easier to fill the web with LLM-generated dreck, it\u2019s to help me (and hopefully you, too!) improve your writing, while still keeping it fundamentally yours.<\/p>\n

Where Does Creativity Live?<\/h2>\n

When you\u2019re evaluating these tools, \u201ccan an LLM do this?\u201d isn\u2019t the question you need to ask. Instead, think about where the creative part of the process lives. For software, that\u2019s in the design decisions and the architecture, the final product is the expression of that creativity. The specifics of the implementation don\u2019t really matter. For a blog post, or any writing for that matter, the creativity lives in the act of writing. To delegate that to an LLM is to delegate your own creativity.<\/p>\n

Here\u2019s what I believe: the best uses of LLM tools are when they augment humans, rather than try to replace them. They enhance the inherent creativity of their human operator, they don\u2019t suppress it.<\/p>\n

This belief guides how I use LLMs, and how I build tools that help others use LLMs, too. I\u2019ll be pushing out a new release of Claudaborative Editing in the next few days, I hope you\u2019ll give it a go!<\/p>","protected":false},"excerpt":{"rendered":"

If you\u2019ve been paying attention to LLM-based coding tools in the past few months, you\u2019ll have seen a seismic shift in how they\u2019re being used. Even 12 months ago, they were little more than glorified auto-complete tools: useful for quickly repeating patterns, but terrible at producing well structured, thoughtful, maintainable code. More recently, however, there […]<\/p>\n","protected":false},"author":1,"featured_media":4642,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"advanced_seo_description":"","jetpack_seo_html_title":"","jetpack_seo_noindex":false,"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1],"tags":[],"class_list":["post-4641","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-latest-news"],"jetpack_featured_media_url":"https:\/\/mlcqvjhyzqda.i.optimole.com\/cb:K5rb.7cb\/w:auto\/h:auto\/q:mauto\/f:best\/https:\/\/i0.wp.com\/frontlinenewsng.org\/wp-content\/uploads\/2026\/04\/Screenshot-2026-04-07-at-3.01.26-pm-333x1024-4DKYJG.png?fit=333%2C1024&ssl=1","jetpack_likes_enabled":true,"jetpack-related-posts":[],"jetpack_sharing_enabled":true,"_links":{"self":[{"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=\/wp\/v2\/posts\/4641","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=4641"}],"version-history":[{"count":0,"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=\/wp\/v2\/posts\/4641\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=\/wp\/v2\/media\/4642"}],"wp:attachment":[{"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=4641"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=4641"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/frontlinenewsng.org\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=4641"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}