[I work on Claude Code] I broadly agree with the author’s point: plan mode was useful, and is no longer useful.
In Claude Code, all plan mode does is add a little reminder to every user message along the lines of “you’re in plan mode, please don’t code yet”. It’s something I came up with late on a Sunday night many months ago, when I got tired of asking Claude to plan with me first before coding in each new session. Something people might not realize is plan mode has always been a prompt — it has never changed the toolset because doing so would break the prompt cache, and so would be expensive for users.
This worked well for a while, until a few months ago, using early versions of Fable, I realized that I wasn’t using plan mode anymore because the model just got it, and because for the increasingly complex work I asked the model to do, planning had become interactive and iterative. With Opus 5.5, I feel Opus has gotten to that point too.
For codebase understanding, I sometimes ask Claude to generate an artifact that explains some aspect of its changes. For complex diffs to core parts of the system, I will often ask it to make diagrams or even interactive demos so I can better understand the change and alternatives considered. I don’t do this very often, but it’s a useful way to explain code when you need it. I ask Claude to attach these artifacts to its PRs also, so others can understand and future Claudes have the context.
Any move away from the abusive big tech that exists in USA is a good move in my book. Microsoft just recently made a new patent where they will monitor you via camera and microphone to block your program/game to show you ads, and per ads viewed, you then earn credit to be able to use your program/game. It is absolutely wild to me how a corporation can be so toxic and abusive to its users and still be in business.
Hi, your friendly, local pathologist here. I'm the kind of doctor that diagnoses liver diseases, cancer, lots of infections, inflammatory processes, benign tumors, and runs the lab that a lot of other doctors use the results from to inform a lot of their other decisions.
If you really want to learn about this stuff, Robert Weinberg's The Biology of Cancer is where to start. It introduces concepts starting where a high school student could understand and carefully walks through the history of our development of knowledge until, by the end, the pithy chapters are helping novel concepts for real science flower in your brain all on their own. It's the best book of biological science I have ever read. 10/10, will read again.
It looks like a primitive chess engine, trying every move, no matter how stupid, until it works. Relying on its ability to do millions of operations rather than having a plan.
People will try stuff too, but once there is an opening, they will consolidate, generalize, simplify,... before going to the next step. The agents didn't, it is a huge, vaguely directed mess.
Also, it looked so "loud", querying millions of URL with weird requests. The sandbox as weak as it can get, and there is absolutely zero smart extrusion detection or it would have found it. They used their best AI for attacking, but nothing for protection.
A certain nuclear power plant had a Windows NT 4.0 machine running as late as 2007. The reason is interesting.
The machine's purpose was to report status of the control rods that mitigate nuclear reactions. Basically, "are the rods inserted, and if so, how many / how far?". I want to emphasize that this was reporting only, NOT control.
The original software was written back in the 80's, when the plant was originally commissioned, for AmigaOS. Of course, it's hard to buy Amigas anymore, and the original one died long ago (nobody remembers when).
So in the mid '90s, the utility purchased an AmigaOS emulator that ran on Windows NT 4.0, which was current at the time. The emulator (IIRC) was developed by a firm in the UK. The firm went out of business sometime in the late '90s. The control rod monitoring software ran under this emulator on top of NT4.
Windows NT 4.0 was the last OS to allow the emulation software direct access to the physical hardware that produced the status signal. Later versions of Windows abstracted the hardware access away, and the monitoring software broke. Because the emulation company had gone belly up, there was no way to fix the incompatibility.
So the utility had a choice: get new hardware/software certified (by NRC?), or keep doing what they were doing with the software (and hardware) that they had. They chose the latter.
So this is how, in 2007, during a tour of the facility, I stumbled across a Pentium 1 system running an AmigaOS emulator on Windows NT 4.0 that was responsible for displaying the status of the control rods of a nuclear power plant.
Spare hardware for this setup was purchased off of eBay and stocked on an adjacent shelf.
I'm actively watching understanding slip away from developers, code review getting paired down to no comment checkmarks, and codebases go to bloated messes that nobody can read. Axioms like engineers must understand and take responsibility for the code they ship are getting torn down, and the products coming out are reflecting conway's law, becoming impenetrably obtuse and always "so complex there are no obvious deficiencies" (as opposed to "so simple there are no obvious deficiencies" which used to be the aim).
The one thing plan mode helped is for the humans to get an understanding of the strategy, and be able to poke around and look at the design and architecture. You can achieve this with some self discipline and keeping shorter leashes on agents, but it feels like a losing battle. The best devs still put out good code, but the poor devs are learning nothing while their metrics look great. I can't help but think we are racking up immense amounts of debt that will very soon become due.
My biggest takeaway from this is just how godawful the sandboxing is. The stuff written up in OpenAIs report says more about lack of extremely basic sysadmin skills than anything else.
I’m not that surprised about models with endless compute being capable of this, I’m more surprised that a company with the resources they have apparently can only create a sandbox that a half skilled human operator could have broken out of easily.
I think it bothers OP less that they take a 15% tax than the fact that google provides terrible support for their own play store. If they would take that tax and provide good feedback and speedy version reviews, nobody would ever complain - it is expected to pay something since the play store doesn't run on good thoughts and prayers.
But because they're a monopoly (or a duopoly if you count apple, which is a different platform altogether) they can afford to act this way.
Yes, the physics are worse, and yes, the economics are worse, but data centers in space have the crucial advantage that they are out of range of the molotov-throwing arm of Joe Public (recently unemployed).
> Before approving construction, I would want communities of humans to understand why the design works and what justifies confidence in its safety. I would hope that we all would.
Until very recently, I pored over every single line of code Claude generated with razor sharp scrutiny. I would usually catch issues with every response. I'm catching fewer problems these days. Maybe the model is just getting better, and maybe I'm being less careful while under pressure to ship more and more often. But model capability is obviously growing. Even back in March, you could tell it "give me a function that adds two numbers" and you could be 100% confident that it would write the correct function. There was almost no point in looking at the code. Since then, the complexity floor of problems in the category "this is so simple that the model couldn't possibly get it wrong" is rising, and with it, my cognitive surrender to the model is increasing too. Why check it? It's obviously going to be correct.
If AI designs a terawatt fusion plant, then of course we're going to meticulously pore over every detail to ensure safety, reliability, efficiency, whatever. If we find no flaws in the design whatsoever, will we be less careful about the second one? The third one? What about the ten thousandth one? Will "a nuclear fusion plant" become something that models couldn't possibly get wrong?
Terence Tao is arguing that the human involvement in research is crucial, but doesn't convincingly justify why, in my opinion. He says that "human agency is a value of fundamental importance" and that we will need to build "thriving human communities that can understand [AI ideas] together" - not for the sake of correctness, which AI may surpass us on, but for, I guess, the possibility of reclaiming human meaning and purpose. I don't disagree with this at all, but it's not an argument, it's a statement of values. Unfortunately, the stark reality is that if AI does surpass humans, it will become the economically dominant strategy to not verify them and not double check them, but to just do whatever they say. This seems like a great way to raise p(doom). But as the models get better and better, and as I'm scrutinizing Claude's output less and less... I just hope that there are more Terence Taos out there than people like me.
Explainer videos are dark patterns that replaced written howto documents in order to serve ads. Not seeing them pop up any more in search results is one of the few positive outcomes of Google switching to AI-first search results.
Imagine having a virus escape a sandbox, why are we worried about the virus but not the incompetency of those who are responsible for setting up the sandbox?
If I post something on the Internet today claiming that I asked my agent to do X but it went rogue and did Y, all I will be getting in return is a jar full of "skill issue".
Should we worried about people using LLMs for attacks? Yes, but not in the premise of LLMs going rogue but someone with the intention of abusing it to cause harm. And this is not something we as individual or even company can deal with, responsibility should be held by those who use it, in a legal way.
I am baffled by the fact that up until now, no one is held responsible for so many incidents reported publicly or privately. At this point, it's free marketing, if I am CEO of any AI company, I will run swarm of agents hacking all NGOs and stating that I am just looking for some random piece of data that happened to be hidden in their servers, at least that's what my LLMs think, not me. Then I will start preaching everyone how dangerous this piece of technology is and start giving out free tokens for these NGOs so they can start defending themselves and we should slow the f down.
Getting bored of these framings where the superintelligent sentient beings running freely inside OpenAI are doing things that the company has no control over. The headline should be:
OpenAI meddled with multiple US Government agency sites.
The bots are acting neither properly nor improperly, they’re acting as they’re being allowed or coordinated to act.
I know everyone says this is political but it actually seems like a textbook designation. Anthropic wanted to have rules on how the military used AI, the military said no and therefore doesn't want anthropic used anywhere in their supply line.
This is like a pen manufacturer not wanting their pens used to sign drone strike orders, now the military needs to have a special box of pens that don't have stipulations attached. With AI usage it would be the same thing except applied to entire product chains. It seems like it would just add more complexity to operations.
You can agree with the rules anthropic wanted, but having rules set by a private company at all that apply to the military does seem fair for the military to object to.
>The Department reasonably feared that Anthropic might manipulate Claude’s design to prevent it from performing national-security functions that the Department deems contractually authorized and necessary
Though they'd probably put the DoD on the cybersecurity whitelist today, the very idea of the claude whitelists for certain functionality already exists and is being used by them today.
Not sure if I’m just in my own national security bubble, but I find it troubling to see the perspective that most of the conversation in this thread is coming from.
The US government took a legal designation explicitly crafted to protect against foreign adversaries and deployed it against a private, domestic entity, to their immediate and great detriment.
I think the better analogy is an insane nuclear power plant manager deciding it wants to buy pens to use as neutron-flux regulator rods — because after all, a pen is functionally a pencil and a pencil is made of graphite.
Then the pen manufacturer hears about this and says “Our pens are not made of graphite and are not suitable to be used in nuclear reactors”, to which the reactor owner says “it’s fine, they fit in the graphite rod holes, and we’re just using until the next generation of pens come out which will do an even better job”, and then the pen manufacturer says “I’m not going to sell you any pens until you agree that they will be used only writing.”
This entire conversation around Jev seems weird to me. Like... we started from neural nets that could do basic decision making and classifications pretty well, then trained larger and larger language models to get to where we are now. Now suddenly everyone is going crazy because someone trained a smaller model that is adequate at making decisions? We already went through the "look this AI can play pokemon terribly" phase like a decade ago.
Hey, all. I really don't know what to do with a post like this.
I'm being sincere when I say (as I've said on two threads here) that this genre of posts --- "I'm leaving this company I've been very publicly associated with, and here's the new thing I'm doing" --- is deeply cursed. There's no way to say anything interesting without it just stinking like an ad for the new thing.
Obviously, anything at all you say about a commercial project you're working on is easily read as promotional. And you're right, this kind of writing almost always is promotional. But there's a way to do it where at least you're trying to be in conversation with your peers, rather than hitting people over the head with how awesome you think the project is.
But I don't know how to do that in a post like this. I think the only way to read it is as, like, an investor memo. Not my goal, but I don't make the rules.
So my strategy here is just to stay kind of vague, and talk about where I think the world is going, rather than the specific thing we're doing. I can talk your ears off about capability systems, datalog, models driving hardware, virtualization, whatever. Those are fun conversations and I'm very psyched to have them; it's what lights me up about the work we're doing now.
But I don't think it can work here. I didn't submit this post and I didn't upvote it. I wrote it because I didn't want the whole thing I'm leaving Fly.io for to be wrapped up in some dumb Twitter thread.
If you're unsatisfied with the post, I don't blame you, but it's less a bid for the front page of HN than it is an update to my "about me" page. I'd literally rather talk about HN meta, and how to write for HN, than I would about operating systems at this moment. I truly appreciate the interest though.
It's useful because it let's me see the decisions the model will make before it wastes a ton of time implementing them. The model is smarter now but that doesn't solve for underspecification if it guesses my intent wrong
> This worked well for a while, until a few months ago, using early versions of Fable, I realized that I wasn’t using plan mode anymore because the model just got it, and because for the increasingly complex work I asked the model to do, planning had become interactive and iterative. With Opus 5.5, I feel Opus has gotten to that point too.
For me it’s actually the opposite, and Claude Code’s plan mode isn’t nearly sufficient. Personally I ask Claude to write down a markdown file with its plan, then review the plan using plannotator, and then go back and forth (most of the time it’s actually the comments that are the problem, not the code).
Then start a fresh session, seed it with the plan, tell Claude to find ambiguities / friction points / oversights, resolve those, and then implement it.
Review once again with plannotator, go back and forth, and then send PR.
Maybe not the “vibe coding” that was once imagined, but this does ensure I am fully aware of the code and architecture, the quality, and this also prevents long term degradation.
Cars that are designed from the ground up to be EVs tend to be much better than trying to shoehorn an EV into an ICE design.
The new Lexus ES350e, which comes in both hybrid and EV flavors, has been pretty universally panned as a mediocre EV [1]. Lexus anticipates that the vast majority of ES sales will be for the hybrid model, and I expect the same will be true of the Corolla.
The updated bZ and new bZ Woodland on the other hand, which are built as ground-up EVs on the e-TNGA platform, actually seem pretty decent. And I'm hopeful for the upcoming EV Highlander, too (although that will use a modified version of the TNGA-K platform).
I would prefer if Toyota just released a new EV model with roughly the same proportions as the Corolla, instead of going with the one-size-fits-all approach. But maybe they'll finally be able to get it right this time.
What I'm taking from this is that no one is happy about anything, ever. New design is fine. Old design is fine. I just want to search for an app and find that app, which is mostly what happens.
I appreciate all the hard work that goes into F-Droid <3
The drop from 2018 to 2022 is as big as the one from 2022 to 2026, so it's not obvious whether AI had a role at all, although it's highly plausible.
I suspect the bigger culprit is the optimized monetization of human attention, because pre-algorithmic social media doesn't seem to be as destructive. The fact that science wasn't as affected as math and reading (both of which rely more on attention/practice than rote memorization) somewhat supports this.
> I know everyone says this is political but it actually seems like a textbook designation
It literally is a textbook definition, signed into US law:
“Supply chain risk,” means the risk that an adversary may sabotage, maliciously introduce unwanted function, or otherwise subvert the design, integrity, manufacturing, production, distribution, installation, operation, or maintenance of a covered system so as to surveil, deny, disrupt, or otherwise degrade the function, use, or operation of such system (see 10 U.S.C. 3252).
To add onto what another commenter said, the pen analogy would be more like the manufacturer designing pens that stopped working when used to sign strike orders they disagreed with.
I think you're confused. When the White House decided to stop having federal agencies buy paper straws [1], they didn't designate paper straws a supply chain risk, they just stopped buying them. The term has a very specific meaning which would not apply in the pen scenario.
Very well written both in prose and tone. I'm glad she decided to tell this story. If the author isn't a professional writer I think she could be.
This is a great "behind the scenes" style look at the actual people behind a viral clip. Luckily they have an extremely strong relationship to help soften the blows here, I'm not sure an average (not even bad!) relationship could come out of something like this unscathed.
There's something feral in us that comes out from time to time, especially online from the safety of our screens. I found the announcers within the range of good fun but when it gets to people reaching out to break up their marriage or telling him to kill himself it becomes really sobering.
I think this is part of what people are calling the lonlieness epidemic - despite thousands of screaming voices the whole thing makes me feel hollow and hopeless.
A really toxic part of online interactions is that it's very common for people to see a tiny snippet of another person's life and then project all of their own personal resentments onto them.
I really dislike attitudes like the author's. I think it's great and important that people and companies spend considerable amounts of their time and money exploring the limits of design and what's possible, even if the concepts produced have little direct, practical use today. That kind of speculative R&D can lead to great things and who knows, maybe in 20 years it will lead to something amazing that will benefit a great deal of people.
Trying to pooh-pooh the ideas of tomorrow because they don't immediately solve all the problems of today is just terrible, short-sighted thinking.
And on a more pedantic note, obviously someone asked for it, else it wouldn't have been made at all. What the author really means is "I didn't ask for a crab chair".
In Claude Code, all plan mode does is add a little reminder to every user message along the lines of “you’re in plan mode, please don’t code yet”. It’s something I came up with late on a Sunday night many months ago, when I got tired of asking Claude to plan with me first before coding in each new session. Something people might not realize is plan mode has always been a prompt — it has never changed the toolset because doing so would break the prompt cache, and so would be expensive for users.
This worked well for a while, until a few months ago, using early versions of Fable, I realized that I wasn’t using plan mode anymore because the model just got it, and because for the increasingly complex work I asked the model to do, planning had become interactive and iterative. With Opus 5.5, I feel Opus has gotten to that point too.
For codebase understanding, I sometimes ask Claude to generate an artifact that explains some aspect of its changes. For complex diffs to core parts of the system, I will often ask it to make diagrams or even interactive demos so I can better understand the change and alternatives considered. I don’t do this very often, but it’s a useful way to explain code when you need it. I ask Claude to attach these artifacts to its PRs also, so others can understand and future Claudes have the context.