Should You Use AI for a Task? Here’s a Simple Way to Decide - Schneier on Security

But it’s precisely because I have spent years developing my own writing skills that I’m able to identify prose that sounds great but doesn’t actually make sense. My students don’t have that skill; they mistakenly view a confident, well-written essay as evidence of the quality of their ideas. They see the AI as cleaning those ideas up, getting them through that uncomfortable stretch of having to turn those ideas into prose. What the students miss is that their initial discomfort is a normal and healthy stage of writing, and not something to quickly get beyond. The very act of struggling with how to express what they think is an important part of the process. It’s how they test out their ideas, examine their hypotheses, and actually figure out what they think. Homework is not work; it’s the gym.

Learning can be hard, for many people at least. I like the feeling of being outside my comfort zone when working on something because I know that at some point, the journey is going to be the reward. We forget that in the age of AI.

Yes this is scary as hell

The real AI risk is inside the labs -

I believe we should not consider AI safe. A critical event that may result in the extinction of Homo sapiens is possible, but the danger is not in open models, or China making faster progress than the US. The danger is that a few CEOs (everywhere in the world) without the required background and legitimacy are in the position of making hard choices for humanity at large. They were not selected to do so; it was just the randomness of events that created this setup. They can’t speak for everybody, given the stakes, just because they have GPUs and money. This is the first thing that should be fixed.

In light of recent security-related events, both at OpenAI and Anthropic, I also believe these companies must be held accountable for what they put out there. I believe we are entering dangerous times with AI, maybe we are even on the verge of being out of control with this. Add this to the AI bubble speculation. Scary.

Too much to ask

For the past two or three years, I’ve been contemplating the idea of creating a comprehensive training course for Craft. There’s nothing comparable to what you find for Notion or Obsidian. I think it’s a great opportunity, very niche, but still an opportunity. Craft has evolved considerably in recent years, making it difficult to produce content that doesn’t become outdated quickly. Now that Craft’s evolution seems to be slowing down, the idea of creating this training keeps resurfacing. But there’s a problem: a lack of motivation and time with no guarantee of results.

I tested creating a training plan using Claude AI, and while I’m impressed by the level of detail and completeness of the proposed content, I would need to record more than 75 video capsules lasting 2 to 4 minutes each. Preparing those recordings requires between 15 to 45 minutes of preparation, without considering the post-processing requirements. That’s quite a commitment!

I’m afraid the idea won’t materialize. Life is short and I’m not here on a humanitarian mission. I think I’ll move on to something else. 😌

WTF Substack? What are you for?

I’m unsure who Substack writers aim to reach. In my view, Substack might be the most anti-reader platform I’ve ever seen, even more so than Medium, which is in itself very telling. Today, I wanted to read an article on Substack after following a link in an email from my inbox, but my app wasn’t installed on my iPhone. After reinstalling it, I tried to open the link again, but it didn’t work. I had to log back in, apparently, which I did. The next step? Reading the article? Not quite—now it asks for topics of interest, then requests access to my Contacts. I stopped there, killed and removed the app from my phone and never read that article. Instead, I’m here to vent my frustration.

Links to Claude Chats Were Indexed by Google and ‘Bing’, Whatever That Is — Pixel Envy

Search engines do not automatically discover new URLs as they are created. They are typically indexed by taking one of two actions: either the site owner submits URLs to search engines — by using a site map, for example, or by manually submitting the page for indexing — or by someone linking to the URL from a different already-indexed page. The critical thing left unreported in Varner’s article is whether these Claude links were being indexed thanks to Anthropic’s actions or those of users.

Following up on my previous blog post, I’m now confident my private chat sessions are “safe”.

About a potential AI bubble

I follow technological news every day, and I’m starting to get nervous about the situation surrounding investments in artificial intelligence infrastructure. This video is particularly revealing and convincing. There’s increasing talk of a tech bubble, much bigger than the dot-com bubble in the early 2000s. What worries me, actually, is the impact of this crisis on my finances. I’ll be retiring in less than 7 years. A financial crisis of the magnitude being predicted would take years to recover from. I wonder if it’s time to reevaluate some of my investment strategies. 🤨

I don’t want to link my WhatsApp account with my Facebook account. I don’t use Facebook. Inuse WhatsApp mostly while abroad during vacation. I know I would make a favour to Meta if I did link both in the name of convenience. I’m probably being naive here, though. Postponed indefinitely.

About Glaze

I played a little bit with Glaze today. In less than half an hour, I was able to build a “native” macOS app, a front end to my web-based task manager. It’s far from north stars like Things 3, but it is quite capable for the efforts I put into this experiment. Glaze offers a Glaze Store where app builders can share their work with others. A brand new account comes with 120 free credits. I’m currently at a bit less than a hundred credits left now after building this little Mac app. It’s currently in read-only mode and looks very dry but I plan to tweak it and see what I can come up with the remaining credits.

Ship it when it's ready

MacRumors’s article “Anthropic’s New Claude Opus 5 Nearly Matches Flagship Fable 5 at Half the Cost”:

Anthropic has released Claude Opus 5, an upgraded AI model that delivers near-Fable 5 performance at half the cost and is optimized for everyday use. The new model shows significant improvements over its predecessor (Opus 4.8) across multiple benchmarks including coding, problem-solving, and visual outputs, while maintaining the same pricing structure. Opus 5 is now available as the default model on Claude Max and the strongest option on Claude Pro, with an optional faster mode available at double the cost.

It’s fascinating to watch AI companies release their new offerings. It’s all over the place if you compare that to the way Apple releases their flagship products like the iPhone. “Ship it as soon as it’s ready” seems to be the way to go.