When I was 5 years old, my Granny and Granddaddy Hodge took me to Disney World. It’s one of the earliest memories I have. I remember sitting on the curb on Main Street, U.S.A, watching with wonder and pure happiness as the parade passed. I can remember how hot it was. I don’t think the air conditioning worked in my Granddaddy’s car. We stopped on the way back to Alabama to trade in that car. My grandparents loved to tell this story. Little 5 year old me said to my grandfather, in front of the salesman as if my grandfather had something to hide, “Don’t forget to tell him about the air, Granddaddy.” This was a magical trip for me and formative in how I think about entertainment and business.
I was living in Mississippi at that age, so it was a long trip for me. After we made it back to Alabama, I’m sure we rested up at my grandparents house before they would have had to make the trip back to Mississippi, in that new car, to return me to my mom. I was lonely in Mississippi even at 5 years old. I had no family there. My mom worked during the day and spent her evenings with random men I never really met. I spent my days in daycare and my nights with comic books and action figures, trying to image a better world for myself. Disney World with my grandparents was that better world, a place where I could be safe and sing along with everything enormous and magical going on around me.
A few years later, my mom and I moved back to Alabama. We were getting threatening calls at our apartment, which I understand now were probably from one of those men I never met. This terrified my mom, and when we got back to Alabama, she had her very own come to Jesus moment. It didn’t last that long for her, but the lasting effect for me was that I ended up in the ultra-conservative, independent Baptist, private Christian school run by my Grandmother Catchings’ church.
Normal Southern Baptists are hellfire and brimstone people. Independent Baptists are all that and more. Women couldn’t wear pants. Men and boys alike couldn’t wear shorts in PE class. I had to wear a tie to school. My hair had to be cut with what they called “white walls” above my ears and collar. I attended this school from 10 years old until I was 14. At some point in all that, I had to take a class called “Man in Demand.” It was taught by the preacher of that church. In class one day, the preacher told me that I would never be a real man because I was being raised at home by a single mom. Never mind that I saw my dad regularly throughout the year.
This preacher who told me this was the second preacher to run this church while I was there. When he joined, he was welcomed as a softer, more empathetic pastor for this congregation. That’s independent Baptists for you.
This kind of thing happened often at that church and school. I’m not sure what year it was but somewhere in there, there was talk of going back to Disney World. My cousins on the Catchings side lived in Orlando, and we were talking about a visit. I was hoping we could go to Disney World. I mentioned this in school, probably unable to contain my excitement about escaping that terrible place, and one of the teachers told me that the church didn’t support going to Disney World. Apparently, it was a place of filth designed to indoctrinate young children into the ways of Satan. I was like, “but have you been there?” Then I spent the afternoon in detention for smarting off to the teachers.
It’s a fair question, though, even if I didn’t know it then.
I’m sure none of the teachers working at that school could have afforded to go to Disney World. I wouldn’t have been able to either if not for my grandparents Hodge. They saved their money well while working at the local textile mill. They invested in the stock market. They started their own rental property business in that small Alabama city. It allowed them to travel. It allowed me to travel with them. It allowed my dad to get an education and get out of that city, which ultimately did the same for me.
It’s easy to be against something you’ve never experienced, and it’s easy to make commerce the villain if commerce isn’t working for you. That’s really what religious boycotts are all about. It’s about that lack of access and all the feelings that go along with that: frustration, resentment, anger, and fear. It’s the emotions that drive the boycott, not any sort of rational thinking. It’s all the same—whether Target or Tesla, DEI policies or Harry Potter, Disney World or Artificial Intelligence—where one person sees a religious war to wage, there’s a 5 year old in Mississippi, or a 10 year old in Alabama, who sees opportunity and the possibility of a better world.
I’m surprised by how many people in my circle aren’t aware of the great deal that is the Apple News+ subscription. Yeah, it’s $12.99 per month, which might seem like a lot for a news subscription, but it has literally 100s of subscriptions—both magazines and newspapers—that would all be hidden behind their own subscription fees. It’s actually a great value for what you get.
You can find 3 months free for new subscribers at Apple’s own News+ site. Lots of other places online offer bundled subscriptions too, especially with new device or new service purchases. Look around for deals. They’re out there.
I’m reading The New Yorker, Wired, The Atlantic, as well as The Wall Street Journal, The Washington Post, Los Angeles Times, and The Times of London. Those are all subscription based. I wouldn’t buy all of those individually, so Apple News+ is bringing me a lot of value.
Some clarifications and disclosures: I have no idea if this is a good deal for the news organizations themselves. I understand why a news site like The New York Times or CNN would want a direct to consumer relationship, rather than being mediated through Apple News. I work as a software developer for CNN but don't really work on or have insight into this part of our business.
All that aside, as a news and journalism fan, it’s an incredibly good deal. Check out Apple News+ if you haven’t already!
I wrote previously about the need to take a more skeptical view of Anthropic’s claims about Claude Mythos, its new model that is being released only in preview mode with select Anthropic partners. We’re now starting to get some information from those partners. Results are about what I would have expected. They’re certainly not anywhere near the level of world-ending AI that the “our model is too dangerous to release” approach implies.
Simon Willison—a software developer and blogger whose recent focus has been on generative-AI assisted coding—shared a couple links on this topic, which included a link to Our evaluation of Claude Mythos Preview’s cyber capabilities from the AI Security Institute. This is a research organization within the UK government's Department for Science, Innovation and Technology, so it’s an impartial evaluation I would trust.
The AISI describes their results:
Mythos Preview’s success on one cyber range indicates that is at least capable of autonomously attacking small, weakly defended and vulnerable enterprise systems where access to a network has been gained. However, our ranges have important differences from real-world environments that make them easier targets. They lack security features that are often present, such as active defenders and defensive tooling. There are also no penalties for the model for undertaking actions that would trigger security alerts. This means we cannot say for sure whether Mythos Preview would be able to attack well-defended systems.
They conclude:
Our testing shows that Mythos Preview can exploit systems with weak security posture, and it is likely that more models with these capabilities will be developed. This highlights the importance of cybersecurity basics, such as regular application of security updates, robust access controls, security configuration, and comprehensive logging.
Again, this feels normal to me, not software-world shattering. It seems Claude Mythos is indeed better at finding and using exploits, but these tests are being done in simulated environments. Criminal hackers could almost certainly use they to some success, but it doesn’t seem it would lead to anything approaching a shut-down-the-Internet level of security exploits.
I don’t mean to say we shouldn’t take these capabilities seriously. Of course, we should. This is impressive, technically speaking. Model capabilities are clearly improving, and software developers and technologists should understand the implications of those improvements. We should just be careful to not give Anthropic’s hyperbolic claims—claims meant to gin up fear, corporate interest, and spending on their services—any more weight than they deserve.
“I have two really big concerns,” she said. “One, that it’s being used to create fake information that’s destroying our whole information environment. And two, people are trusting them to do things that they shouldn’t be trusted to do. We overestimate their capabilities. There’s a lot of magical thinking about AI. But it must be said that if you let these systems loose in the real world and they have access to your bank account, even if they’re just role-playing, it could still have catastrophic effects.”
The emphasis above is mine. Calling content generated by an LLM “AI” was a marketing stroke of genius. The problem is people have bought into the fiction. That’s the power of a good story.
It might be near impossible to turn back from this perception and belief now, but we have to try. We need an even more powerful story. I appreciate that Gefter is trying.
The big stuff wasnt what I expected. I thought I'd find wrong logic or security holes. Instead what I found was this layer of... I dont even know what to call it. Plausible-looking code that technically works but is structured in ways no human would choose. A try-catch around a console.log. A utility function that was 40 lines of enterprise-grade typescript doing exactly what Array.prototype.map already does. Variable names that sound right but dont match what the variable actually holds, like a thing called userPreferences that was actually a session token.
Tomorrow marks my 1 year anniversary working for CNN at Warner Bros. Discovery. The year has flown by, which I think is a good sign for a job. My last couple roles before this one didn’t work out as well. Every day of those years felt like work. Hard work! So it’s a nice change of pace to look up and go, Oh hey, I’ve been here a year.
That’s not to say we don’t work hard at CNN. We launched our CNN All Access subscription last year, doing it in record time for what that build out required. There were other major internal initiatives that my team directly contributed to, but it has been an enjoyable kind of hard work. I find it usually is when you’re working with people you respect and in a healthy environment. One thing I really love about media companies is that media companies prioritize humanity in work. That focus on people coupled with interesting software engineering is the secret to a great role for me.
CNN has certainly been all of that so far! I'm super excited to celebrate my 1 year anniversary. Hopefully it’s just the first of many anniversaries to come.
I’ve had this experience a lot lately at work—someone comes up to me, they’re smiling, they’ve got news to share. “Have you seen what these things can do?” they ask. I don’t want to be a jerk, but in my heart and head, I’m like, “Why yes, actually, I have.” They’ve just discovered what generative AI can do. Their minds are blown.
Anyone who’s done anything more than chat with a chat bot has had this experience. I made a comic book prototype with it. I’ve used it to code. I’ve run simulated fantasy football games with it. One of my daughters has tried letting it make “cringy fan fiction,” as she calls it. It’s all pretty magical. Until it isn’t.
Anyone who tells you about generative AI with that twinkle in their eye, like they’ve discovered the old magic, just hasn’t used it long enough yet. I want to say, “Just wait. You’ll find its limits, and then you’ll come back down to earth with the rest of us.”
I’m really curious to see what happens when the rest of the world reaches this point.
Who am I even writing this thing for? I have some ideas, but I feel the need to write them down to see what I really think. So here we go.
A General Audience Interested in Technology, Media, and Writing
My impulse is to say I’m writing for a general audience with an interest in technology, media, and writing. I think that’s mostly true, but practically, it’s not really how I’m writing things here, where “things” means blog posts. My short stories are definitely written with a general audience in mind.
I know I’m currently using too much industry jargon and assume too often that people reading my site are computer nerds like me or writers interested in writing about technology. If I’m going to welcome a more general audience to this site, I need to get better at tweaking my writing for that audience. I’ve got a longer piece I’m working on that I’m beginning to tweak with this audience in mind.
Let’s see how that piece turns out.
Documenting the Journey
I’m also writing this post because I want to do more thinking out loud with my audience here. I wrote previously about my goal of writing 1000 words a day. I also want to be transparent about this goal of building an audience and documenting my process. The best sites online let readers into the lives of the person behind the site. I’ve been blogging or maintaining my own personal site off and on for 20 years now. I’ve never really tried to build an audience or a relationship with that audience. I’d like to take a stab at that and see how it turns out.
Why?
Well… why not? But more seriously…
I found my way into web development quite by accident, thinking I would use it as a way of sharing my writing online. In the 90s, when I was in college and discovering the web, I was really into reading work posted online from professors I followed. These were people posting their stuff online—there was a term we used for it then before “digital humanities;” anyone remember that term—and I imagined a similar life and work for myself. Then, I got good at web development, it lead to a fulfilling career, and I never regularly put my writing online.
I’ve still been writing the whole time, just for myself. I’ve also worked for a one or two companies that had a focus on writing. And now, another one. It’s a recurring theme in my career!
A couple years ago, I got the itch to go to grad school and find a better way to merge these two passions. Grad school didn’t work out, but the desire to merge these passions never left. I’ve been trying to find a way back to those heady days of the early web for awhile now.
So that’s why.
Maybe I’m too idealistic. I keep thinking if I could really pair this web and writing thing, there could be a real audience for it. That’s why I’m writing so much now and putting so much effort into this site. I'm trying to take the long view this time. Keep writing, keep posting, and keep tweaking this site until it catches on with the audience it was always meant to find.
Simon Willison posted his thoughts and support for Anthropic restricting Claude Mythos to security researchers. I have immense respect for Simon, for his history in our industry, for helping create Django, and for the work he’s doing to truly understand large language models and their impact on coding, but I wish he would have take even the slightest bit of skepticism with Anthropic’s claims. He even dismisses the need for skepticism outright:
Saying “our model is too dangerous to release” is a great way to build buzz around a new model, but in this case I expect their caution is warranted.
The “our model is too dangerous” thing is not just a way to build buzz, it’s Anthropic’s entire raison d'être. Dario Amodei in particular spends an enormous amount of time out here on the interwebs talking about AI danger. Here’s a video where he basically says if we don’t stand in the gap to save humanity, no one will.
Everyone at OpenAI is so concerned, framed by their OpenAI themed glassware
When I was in my freshman-level college classes working on my liberal arts degree, one of the first things we were taught was to examine the source of a claim. If the person or organization behind a publication has an explicit bias or an outcome they desire, then you need to bring extra scrutiny to those claims. Anthropic and OpenAI not only want these doom-and-gloom scenarios to be true, they need them to be true. Otherwise, they’re just working on normal technology.
I updated my stories page with a new short story. This was a story I wrote in late 2024, which was the last year we did our family writing contest. I've written before about the family writing content.
Here's the opening from that story.
Winter.
The wall calendar was barely hanging on, but it was stuck there, on the fridge, dangling from an ALFA Insurance magnet attached at the corner. The magnet had the too-smiley face of their family insurance salesman on one side, and on the other side, the days of the month from a month that had long-since passed. Brandon Winter was angry his mom kept this mix of calendars on the fridge, past and present pulling at each other and yet somehow working together. He was staring at the date on the wall calendar. It was February 10, two months to the day from when he had been laid off from his job at World Wide Solutions.
