Not going to happen. Most decisions in tech do not factor in risk, only cost. Risk mitigation usually requires considerable spend, and very few people/companies are willing to spend on low-probability events.
Risk is always reducible to cost though. Even for human life, the cost of safety equipment is weighted against the statistical value/cost of a life, instead of spend on for instance maritime equipment being engineered to be infinite. One could argue that the cost is harder to define for squishier subjects, for instance like the privacy of pictures from my vacation or that I send to my physician. But I’d argue that is even simpler, the value is just what one is willing to spend to not risk it, plus any cost of cleanup/being made whole.
I know you're sort of speaking abstractly, and I think you're correct that this is how companies interpret risk, but it's interesting to give a moment's thought to -- Murphy's Law tells us all events have a 100% probability of occurring given a certain amount of time. And we tend to see that proven out time and again.
This might change as AI-assisted hacks and outages become more frequent. A materialized risk with associated costs is very different from an abstract risk that's hard to quantify.
> The same Elon who keeps promising things his companies cannot deliver and obviously exerts incredible pressure on his underlings?
Failing to deliver a ReactJS + web services project, and failing to catch a steel cylinder weighing thousands of tonnes, full of combustible liquids, hurtling down from space with two metal arms mounted on a pole are two different things.
If you're a billionaire and not failing enough times doing absolutely crazy things, that's a net loss for humanity. I want billionaires to fail 80% of the time. Heck even 100x millionaires should be failing a lot.
I'd rather billionaires not exist at all, personally. Before you say something absurd like "SpaceX wouldn't exist without billionaires" I suggest you check the timeline. Falcon 9 was operational before Elon Musk officially joined the 10 digits club.
Innovation will always happen and Elon didn't personally inject billions into SpaceX either.
Let's cap wealth at 999M and anything over the gets redistributed to poor entrepreneurs.
When was Elon promising ReactJS projects? He promises cars, full self driving, etc. etc. Extremely consistently overpromises and underdeliver.
Anyway, "I want billionaires to fail" is fine, cool, I probably agree, but the question being argued was "is Elon running things inherently safer than the situation that led to Challenger etc.". I do not agree with the idea that Elon's companies have a culture where it is safe and repercussion-free to report that things are busted.
If you want the world to convert to EVs, they have to be built for less than people will pay for them. That is the only way for EVs to be sustainable.
Can you name a single western car manufacturer that sells EV at a net profit (considering only the R&D and SG&A attributable to EVs for the non-pure-play companies)? AFAIK, it's only Tesla and BYD today. I'd say Elon has delivered cars.
If you think that SpaceX is less safe than the Shuttle program, can you explain how Falcon rocket has had over 350 consecutive launches since the last in-flight failure (and only 1 failure out of 688 flights in the last 10 years), yet the Space Shuttle had two loss-of-crew failures in just 135 flights?
It seems to me that observational statistics indicate that it is approximately 10 times safer than the Shuttle. And they did that while reducing the price of a launch (including the price we taxpayers pay for NASA launches) by a factor of 10. Again, I'd call that a check mark.
I was responding to your broader statements, which as stated were not strictly about the original narrow context about Challenger. Every company has pretty much the same or similar repercussions with respect to speaking up/out. Apple, or say Timnit Gebru at Google.
> If you read up in this thread, the argument was that Elon is safer than NASA.
There's nothing to decide either way. There have been failures at his companies, and at the same time 13 crewed missions on SpaceX rockets without human casualties.
What he actually delivers (in space) is closer to what everyone else promises.
Everyone else promises a lot, but the US and EU space sectors seem to only deliver pork barrels since I was a kid.
I suspect we'd (not just US, I'm European) do a lot better if we supported a bunch of plucky space startups rather than the big old space companies, which is how SpaceX got its chance to shine.
Musk may have always dreamed big, but with hindsight I'd say he seems to have disconnected from the possible around 2016 or so. Too big to fail, or something.
Not really, and "better funded and started much earlier" is a red herring given how much material science and CNC has changed everything since then.
Given where it's at right now, Starship compares as a mere rough equal on timelines and actual delivery as the N1, Space Shuttle, and Saturn V. Where it might shine, eventually, could be a lot of things, but right now it's about in line with other similar vehicles, just with more pizzazz. The main thing that's changed? We can 3D print what used to be the expensive part.
The Falcon series did good; but the changes were incremental, standing on the shoulders of what came before, and what it got from that advantage was what the Space Shuttle promised.
I can and will criticise the old space companies for their pork barreling, but Starship currently looks closer wanting in on that, than like the SpaceX of a decade ago which made Falcon the success it is.
Would've liked to see them demo some of the forgotten ideas, like having Starship use evaporative cooling. But physics is physics, and re-entry is much the same pain as it's always been.
The problem is when internal stretch goals get communicated as business deliverables that drive stock price. There needs to be an expectation of good-faith for the latter.
The people who buy stock in Musk's companies are fully aware that promises of tech breakthroughs cannot be tied down to a schedule. Risk is inherent in any purchase of stocks, and risk is priced in to the share price.
If you want guarantees, that's what the bond market is for (and at a suitably low ROI).
This implies such promises should not impact stock price. Price = earnings x ratio and they have a time component. If you’re saying the schedule cannot be known, you’re saying they can’t be baked into the price. It would be like trying to price an option without an expiration date.
And yet Dragon versus Starliner. And Boeing was advocating for crewed return after being one thruster from losing control of Starliner. And they also made a pretty bad airplane. And NASA thought STS was safe and thought it was okay to fly every Artemis mission differently with different equipment even when they put people on it.
> If you're a billionaire and not failing enough times doing absolutely crazy things, that's a net loss for humanity. I want billionaires to fail 80% of the time. Heck even 100x millionaires should be failing a lot.
Ah, yes. Others should do more for society so I can benefit!
You should just get straight to the point: you want higher taxes on those wealthier than you.
This is a net loss for everyone, but especially the gaming community. As a founder, he cared about nvidia's history and where it all began.
As someone who had those AGP cards and an nForce motherboard (which were amazing), can't help but feel sad. Whoever his replacement is will not feel the same about what's a much smaller market.
No he didn't. Nvidia could have easily stopped producing any gaming GPUs and allocated all resources to AI data center chips and sold them at 10x the margin.
The next person might not have the same affinity to gaming as he did. He was there at the beginning. He knew what gaming did for Nvidia.
The next person, if it's an outsider, will have no loyalty to gamers. Anyone who can look at a spreadsheet might axe gaming GPUs - until AI supplies catch up.
1) I use 2x ChatGPT Pro 200, and a Claude Max 200.
2) I use other models via Open Router, but ChatGPT beats them in value (at the moment) thanks to Tibo's resets.
3) I mostly make Claude coordinate a group of Codex agents. They've set up a communication channel between them, and Claude acts as the coordinator and a second verifier. This saves me time because it's easier to talk to one agent than a group of agents. Codex parent session does only "comms" with Claude, and all the other agents run as sub agents.
4) For non coding usecases, I use https://www.ctlflow.com. This lets me run dozens of agents on tasks like equity analysis. The information a swarm of agents can surface from a large data store can be quite surprising.
5) I have tested OpenClaw, but now I just use ctlflow.com or claude + codex.
> 3) I mostly make Claude coordinate a group of Codex agents.
Same here. Are you just using a skill or something else. Beyond direct CLI calls to each other, I’m using contextify to share past sessions between them.
> 4) For non coding usecases, I use https://www.ctlflow.com. This lets me run dozens of agents on tasks like equity analysis. The information a swarm of agents can surface from a large data store can be quite surprising.
Can you tell me more about #4 regarding what your heuristics, methods, and use cases are for equity analysis?
I ask because currently I am building a narrow use case Python app using Claude specifically to filter equities that may have an asymmetric potential to buy LEAP options exceeding 1-2+ years.
I am not a finance/quant bro, but I do want to get better at leveraging AI to be a better investor.
I've tried Chinese models extensively over the last 1 year I've been working full-time on this project [1]. They're borderline useless. 90%+ of my code is written by Codex, and perhaps <10% by Claude and myself.
> Can you share any example prompts where Chinese models aren't good enough?
Nobody is prompting any more, it's more planning, automation, guardrails and reviews. For example, if your tests report that there's a 40% slow down in a particular type of analysis, you have to review possible causes, and approve a plan involving multiple agents to find solutions that won't break 50k tests. And that solution itself is a small part of the work that gets planned in a day.
Codex and Claude are the only tools that work for me, inspite of testing almost everything out there to save costs.
You can't use LLMs without prompting them. But, ok, you use codex and it doesnt feel like you prompt them much, but do you have an example commit from codex which you are sure can't be matched by Chinese models?
Hard to explain this without taking up space on an unrelated thread.
But I'll give you a short example. Let's say you have const x = [1, 2] and we want a policy/solver for the type of x (int32 array, int64 array, or floats etc). We could say we'll infer based on the usage of "x", but it's not free. For example, what if it's passed into a generic function add(...), or if there's reassignment or it's aliased, or if it's being used in a loop counter, or a math operation or a combination of these? This is of course doable, but it requires a sweep of a million lines of code to understand the implications on C#, Rust, Mojo targets, and on the current code and tests.
I haven't tried this specific example with OS models, but this is usually the kind of thing that almost immediately falls short and breaks everything. Codex and Claude will do a decent job of analyzing these and giving me better impact analysis. Codex and Claude will also be wrong many times, but the output is often usable.
Add: When I see OS models coming very close to frontier models in coding benchmarks, I just assume based on my experience that they've been benchmaxxed.
Looking at the comments, I get the feeling that HN has no idea how big this is going to get.
Agents are going to move this project into acceptable usability and performance in 6-12 months, across a broad set of Android phones. You're going to see people loading iOS on $299 phones, with grey zone marketplaces for app packages (like Android has had for ever).
Hackintosh was a small hobbyist community. Phones will be 100x that. There's going to be very little money in it, but I'll be surprised if this doesn't explode.
Even if this takes off in some way, the most meaningful result for an end-user would still be to recreate the experience on Android, instead of trying to keep iOS running on foreign hardware.
Reason: Android is a OS designed to run on as much hardware as possible, iOS is designed to support only a VERY limited set of components, and is increasingly wrapped into a per-component security architecture.
I'm saying the hobbyist community for phones is massive - this is not for regular phone users. And they all have older (but fast enough) phones lying around to try something like this on.
And I'm saying that it's not trivial to run an OS on any hardware. Android is designed for this and well-documented, which makes it look easier. iOS is neither of those.
I know, but looking at how quickly people got it running on a few other phones (I've seen pixel 1, 4 and 6, a samsung, and another), I'm thinking that codex or claude code is exceptionally good at this class of work.
This is incredible work. Thanks for letting us know. The really cool thing about this is that it actually works on phones.
TrungNguyen1909's reverse engineering of A13 bionic booted up modern iOS on non apple hardware for the first time. This opened up a viable path to Apple Silicon hackintoshes as well (eventually), and is just inspiring work. That itself was supported by by the work of Martijn de Vos (devos50) on emulating the 1st-gen iPod Touch, winocm on the early 2010s iEmu project for the Apple A4, and the alephsecurity team on the xnu-qemu-arm64 project for virtualizing 64-bit XNU kernels.
I think it continues as Chefkiss inferno and they seem to be close to getting recent iOS to work [3]. I suppose they could make rapid progress with AI now.
We are currently in the process being fully SOC2 compliant and have gotten pen tested twice from two different companies and have found no issues. You can explore the full extent of our security process and pen test results here : trust.alterauth.com
Separately would say it is a bad assumption to make that a product is vibe coded without trying it out. As mentioned in the post we adopted a lot of security processes as a part of our SDLC including consistent scans for supply chain attacks, and static code scanners (SAST).
> Separately would say it is a bad assumption to make that a product is vibe coded without trying it out.
You've put it behind a very typical AI-generated UI; like the many that come up every day. In your case, it's an auth product and it matters more. Besides, It's difficult to try your product without handing you the keys to the kingdom.
> We are currently in the process being fully SOC2 compliant
You're getting a Type I attestation first, I presume? That is of limited value in itself, and is mostly about getting to Type II. If you already have Type I complted, you could mention that on the website. "Regulated industries" have long sales cycles and evaluations, and you'll find it difficult to start those conversations without a more reassuring website.
We are open sourcing the full platform soon so you will be able to inspect the the code.
Regarding SOC2, in the process of getting SOC2, Type 2 compliant. As you may know there is 3 month waiting period before the auditors give a SOC2 certification so currently in the waiting period for that
Not going to happen. Most decisions in tech do not factor in risk, only cost. Risk mitigation usually requires considerable spend, and very few people/companies are willing to spend on low-probability events.
reply