bradfa

Unless you plan to physically assault it with weapons, it’s extremely unlikely to catch fire.

Literally the point of the linked GitHub repo.

Berget's "Cached input tokens are billed at the same rate as regular input tokens" pricing model is kind of painful for lots of coding use-cases, especially when competitors are offering significant discounts for cached input tokens.

There has been massive innovation and development in apples in the past 20 years. Every apple farm with a clue is growing trees for the past decade that look nothing like the trees of old. They’re growing varieties that didn’t exist 20 years ago. If you’re still seeing Red Delicious anywhere, it’s on the buyers at your local markets that missed the boat.

Just read the terms of service and read this blog post and I think your concern will be addressed.

Arduino have muddied this water now, too, as some newer ones come with a full fledged Qualcomm SOC, multiple GB of RAM, and power consumption which requires quite a beefy heatsink.

These attacks are a very effective sales pitch to everyone who runs an internet facing service to utilize AI tools to secure it sooner rather than later. The cynic in me wonders if the marketing team had any influence over the poorly constructed sandboxes or tasks given to the agent swarms when all this went down…

Yes. The vast majority of a solar panel is glass and metals. Both are easily recyclable.

Distance from the equator helps. As does if it snows where the panels are, then vertical simply not having snow on them plus the reflection of the snow helps quite a bit in winter.

If so, then the class of GPU used here may be significantly higher than mere mortals generally have access to simply due to cost.

Obviously nation states will likely have significantly more resources than this, but this is not script kiddie levels of GPUs.

These seem pretty minor compared to hacking HuggingFace.

This writing style, which reads like an LLM wrote it, is draining to read.

I put up with it when I'm actively using an LLM tool, because I just accept that this is how these tools write, but when reading a blog I expect to read words written in one of the vast varieties of normal human-written blog formats.

As long as you are constantly diligent in staying on top of what each provider is doing that may impact you.

It reads to me an argument for self hosting, maybe a less capable model to deal with smaller compute resources, but when starting to use that model and inference software to build a benchmark you can easily rerun when you change things to observe the impact of the change.

Maybe that’s a similar level of diligence but they feel different to me.

Hand written exams are coming back into fashion. You get a pencil and the exam book. Nothing else. Show your work.

Catching cheaters is still hard but it will keep honest people more honest.

Which NEC commercial display do you have? How do you like it for normal TV duties? What tradeoffs were you willing to make when you bought it compared to the features which were available in consumer-focused "smart" TVs?

Or it’s so they can continue to operate in the EU where this is required.

I don't know, but unless that contract is public, not many people know. I get the impression Oracle got a little bit of market pushback on the risk, as theirs seems to be one of the largest.

It would be nice if the article cited the actual contracts the labs have signed so that others could also read them and draw their own conclusions (maybe it does and I got tired of the slop-like writing style too early?).

Are the contracts actually take-or-pay-style? What are the terms? What are the amounts of compute and money involved for each future time period? What happens if the datacenter costs spiral upwards? What happens to the datacenter investment if the buyer goes bankrupt, goes public, or gets acquired (potentially by the datacenter owner)?

I personally think the datacenter build-outs are going to generally be a big swing and a miss. The problem 1-2 years ago was making models good enough to be useful for a variety of tasks. Now we have that. The next problem is making the models efficient enough to run a profitable business. Recent Chinese lab model releases (because they're already constrained on compute resources) and OpenAI price cuts on Luna seem to indicate that this transition to chasing efficiency has already started.

Your comment specifically compared it to the normal adult, my point is the normal adult who is a parent is very likely doing even more than your example of your kid. Even for adults who aren't in the elite high earner category, heck people who make less money are likely working even more hours per day just to make ends meet.

Are normal adults not "working" in a similar way to your kid for 15+ hours per weekday?

Comparable adult parent things would be up at 6am, prep kids for school day, in the office by 8am, 8 hours of work plus a lunch, departing the office around 5pm, transporting kids to/from some activities, dinner, helping kids with homework, somehow fit in regular shopping needs (food, essentials, etc), maybe some volunteering or hobby things, if lucky in bed by 10pm, and we're easily at 15 hours of "work" on a weekday.

Yes, it's a lot.

This morning I've turned auto mode back on, I'm only using it during Claude's exploration phases of my workflow and then I'm reading the full transcript (which includes some thinking info). So far, this seems to be quite close to what I want, and an improvement over my previous "manual all the time" mode.

Still leaving manual approval for all edits. Combined with reading the full transcript of the exploration, I feel I stay in the loop pretty well in this first test.

I run in manual mode not out of rogue agent or token waste concern but because I want to effectively pair program with the agent. I need to understand what is happening along with the how and why. I’m confident I’m slower than YOLO people but that’s a trade off I’m consciously making so I can stay in the loop to comprehend the codebase.

Your summary approval idea is interesting and feels maybe like a mini plan mode. My biggest frustration with the existing manual approval system is when Claude is exploring it gets tedious to approve each command. Being able to approve a block of commands or a mini plan AND have auto mode audit them for safety would probably be something I would consider for the expiration phase of my Claude use.

Agent was told to hack a thing. It couldn’t directly do that so it interpreted the instructions to mean it should hack everything to try to achieve the goal of hacking the main thing. Seems like a reasonable assumption, although a moral human would have understood the context and first asked if that was really the intent.

The AI companies seem pretty bad at setting up tests. And really good at marketing those failures into spin at how amazing their products are.

And Azure Key Vault mentioned. Not that either one was hacked or exploited but the agents got credentials and used them for something (which doesn’t seem fully disclosed). Given that the agents simply obtained totally allowed credentials, which were improperly protected, I don’t think either cloud provider would consider this a breach of their system. Valid credentials are valid. Customer screwed up protecting the credentials.

A small model would be. I think that’s more the point. It’s definitely not SOTA but it’s fast and energy efficient and local.

If you got the $20 in free credits from Meta for signing up when muse-spark-1.1 was release, please note that there's now small print stating "While using free credits your content may be used for product improvement" which was not present at muse-spark-1.1 launch when the credits were given out.

If you don't mind Meta retaining your data, the "Contributor" pricing is deepseek-v4-flash-level of low, roughly 1/10th normal muse-spark API pricing currently. Attractive if you're OK with them retaining and using your data.

We've gotten pretty good at digging holes in the ground and there's a lot of existing energy down there under our land (and near-by water). So yes, we've certainly had misadventures in the middle east, but we've also embraced the "drill baby drill" mantra of late.

Texas is probably our national leader in terms of deploying solar and wind power, which feels ironic but is totally rational and logical.

What’s wrong with bank accounts for your kids? Sorry, this one doesn’t seem to fit with your other examples.

The most interesting trackers to me, for DIY install in my yard, is a 2-axis tracker. Especially in summer, sunrise and sunset are quite a bit north of east and west, meaning a true south facing fixed set of panels wouldn't even see the sun yet.

Sadly, my investigations a few years back indicated that a suitably strong and robust 2-axis tracker that would withstand our local winds and snow loads is significantly more expensive than just buying more panels and facing some of them east/south/west to accomplish the same thing. Doing this on an existing shed also avoids some local zoning rules vs a dedicated ground mounted system.

The proposals which say no LLM generated code mean just for Debian, if upstream allows it then they will still accept it. It’s just contributions to Debian itself which would prohibit or discourage LLM contributions.