How Claude marks AI-generated content:

When a supported Claude model generates text, it weaves an imperceptible watermark directly into the text itself. You won’t see it, and it doesn’t change the meaning, quality, or readability of Claude’s response.

Because the watermark is part of the text, it will travel with the text when it’s copied and pasted elsewhere, and may persist through some editing. Watermarking will be applied at the model level, which means it will be present no matter which Claude product or surface the text comes from.

Today Anthropic rolled out a new support page aimed at users in the European Union as part of signing the EU AI Act’s Article 50(2) Code of Practice on Transparency of AI-Generated Content. The new watermarking will apply to Claude models launched in the EU on or after August 2, 2026. Existing models are covered by a transition period but Anthropic says they’re working to add marking support to them too.

While I understand the concept of how this can work in theory, how it would actually work in practice is mind boggling. Even more so that this will apply even to API outputs from providers like AWS, Microsoft and Google.

How many characters will you have to use for something to be detectably Claude-marked? If I asked Claude to reply only with a single character, how can we be sure it’s marked? The answer is that we don’t know yet, and perhaps Anthropic has really just shipped an CYA page, especially as it pertains to marking older models.

Everyone seems to have an opinion on the effectiveness of “AI detectors” these days, but it will be interesting to see how well Anthropic’s works. If it’s any good, I expect other large economic powers will come knocking.

But I don’t see how the marking system they describe can be any good. Unless the Claude mark is complete gobbledygook (and they said it won’t be), there’s plausibly already text somewhere that would “match” this mark that wasn’t created by Claude. What’s the point of a mark if it’s more likely to cause confusion than no mark at all?

Imagine an author submits their original work to a publisher and they check it against an official Claude mark detector. Even a work with no use of AI whatsoever could may still not pass 100% because there’s nothing preventing someone from choosing the same words Claude might. And if doing so you choose to believe the author, why even check it in the first place?

So far it sounds like a system that solves for a very specific regulatory requirement, but no real problems. I’d anything, it might actually create more problems.

The case for open models just keeps getting better and better.

Simon Willison recently reshared some advice he gave in an interview:

My number one tip for blogging is to lower your standards! Aim to hit publish while you are still actively unhappy with what you have written, because the only alternative is a huge folder full of drafts and never publishing anything at all.

The number one reason I’ve not published more on my blog was that I kept thinking it still needed something. Moving to Micro.blog is in large part a way for me to lower my own bar, and write more on my blog and less just on social media.

Last week I was talking to my friend Josh about blogging and our conversation was self-inspiring enough that I decided to set up another blog.

Well, now my new blog is my old blog too. For the past few years, I’d been occasionally posting at b10g.xyz using a multiple blog platforms hosted on various cloud services over a span of four years. The version I retired last week was hosted on Cloudflare Pages with a site statically generated using Lume from Markdown files I kept in a private GitHub repo. Over the blog’s life, I’d also used Astro and Next.js on Vercel, hosted the current version for a bit on Deno Deploy, and briefly experimented with Emdash. This final version was what worked the best for me. Moving isn’t an indictment of Lume, Cloudflare, or even using Markdown in Git as a CMS. All of those things are great, except maybe the last one, and I still recommend them wholeheartedly. Just not for me.

I originally wasn’t planning to migrate, because that felt like “fixing” something that wasn’t exactly broken. But something was broken. Even though I’d published more blog posts in the last year than I had in a while, I’d come to realize that my self-managed setup lacked restraint. Any feature or tweak I wanted to add, I could. And with the help of AI agents, I often did. What I ended up with was more than enough to validate my desire to write, but I wasn’t publishing all the things I wanted to write because managing what I’d built took more of the time. I still have drafts in my Notes.app that never saw the light of day because actually publishing involved sitting down at the right machine and doing a few manual steps that I couldn’t easily do on mobile.

I’ve learned too late in my writing journey that wrangling your workflow can will drain your capacity for writing and creating if you let it. I really enjoy building platforms. It’s my day job. But I want writing to be joy, not more work. Micro.blog as a platform is exactly what I want. It’s available anywhere, it helps me syndicate to my various social accounts, it has no ads or built-in analytics, but most importantly… it has constraints I can’t change. I’m so excited I signed up for an annual subscription.

Building was fun, but I think writing will be even more fun.

Josh Gammon:

The industries built to help facilitate third spaces in our communities keep getting bought out and shut down. When I think about my stint at Blockbuster during college, I like to joke that I was one of the last employees. It’s not not true. Blockbuster filed for bankruptcy in 2010 and I was working at one of Tulsa’s last remaining stores until it closed sometime in 2011.

Josh and I have been friends since high school and I remember occasionally going to visit him at his Blockbuster before it closed. Even before they announced its closing, the store felt like a relative everyone expected to die soon but wasn’t comfortable discussing. His store finally closed in April 2011 on the same day as half of the 21 stores in the Tulsa metro area, including the one in Broken Arrow I always went to as a kid. I was sad, but I hadn’t rented anything from a Blockbuster it probably 4 or 5 years.

It’s easy to look back and say Blockbuster should have innovated, and in my MBA program that’s exactly what my textbooks and professors said. They thought Blockbuster should have merged with someone bigger. Or they should get rid of late fees, or offer DVDs by mail, or do monthly subscriptions, or kiosks, or streaming. But Blockbuster actually did every single one of these. The reality is that they did these badly and too-late. Blockbuster did not die of a failure of imagination or even a failure to implement new ideas. Blockbuster died because consumer tastes were changing faster than they could.

Jynn Nelson, writing for The “Inside Rust” Blog:

At the time of writing, there are 1,281 open PRs to rust-lang/rust. This represents a staggering amount of time invested by both authors and reviewers. We have long had the problem that there are more people who want to write code than people willing to review it. With the advent of LLMs, this problem only gets worse.

If you’re technically minded, the policy they agreed to adopt is an interesting read, but the overview has its own summary:

It’s fine to use LLMs to answer questions, analyze, distill, refine, check, suggest, review. But not to create.

Jon Udell:

I dislike the phrase “human in the loop” because it cedes authority to the machines. Let’s flip the narrative. It’s our loop, we work the same way we always have, now we recruit agents to join the team. An agent-assisted process need not be a black box that takes in prompts and emits features.

Jon made me realize how easily I accepted the term “human in the loop” in the first place. Of course we’re in the loop, we made the damn loops! Just saying “human in the loop” implies there might one day be a loop without humans at all, which some people are trying to sell but not many people want.

I tend to agree with Jon that “agent in the loop” is a more appropriate disclosure, but if AI was ever capable of doing what big tech said it could, we might end up saying “agent(s) in the loop.” Not because we rightfully acknowledge and wish to disavow machine-centric language, but rather that we would want to acknowledge when an AI agent was keystone reviewer of a loop. As of right now, humans are still better than AI at a large variety of important review tasks, and until that changes “human in the loop” is the better name because human operators are the most valuable part.

I’ve been a fan of Hank Green, and his brother John, since their “Vlogbrothers” series in YouTube’s early days. As well-meaning and well-spoken nerds, they rapidly assembled a community of like-minded terminally online followers. One channel quickly turned into a network, laying the groundwork for educational franchises like SciShow and Crash Course. But with a big audience and community, they’ve also made good. In 2012, the brothers launched a foundation aligned with their goal to “decrease world suck”, and since then it has granted $17 million to more than 100 charities. Earlier this year, Complexly, the company the Green brothers started to produce educational content for their channels and partners like PBS, became a non-profit after both brothers donated their equity in the company.

Which is why the recent fallout over Hank admitting to using ChatGPT for research feels like such a seismic internet event. Hank Green isn’t a faceless media conglomerate chasing a payout, he’s a beloved creator at the center of the internet’s fiercest debate about automation, scale, and burnout. Hank first addressed the issues in a reply to a tweet, but the original tweet to which Hank replied to was deleted, and later posted a longer Reddit apology, admitting he had fallen into an unhealthy habit of relying heavily on generated notes to keep up with his brutal publishing schedule. He announced a pause on several of his personal projects, conceding that “the level of dopamine I’ve been getting from interacting with LLMs… is not healthy for me or good for the world.”