frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

Ask HN: Are we giving away our civilization to the frontier labs?

1•bigwheels•3m ago•1 comments

Robotaxi – Safety

https://www.tesla.com/robotaxi/safety
1•shekhar101•11m ago•0 comments

Mark Zuckerberg Agrees to Stay in California Despite Billionaires' Tax Measure

https://www.wired.com/story/mark-zuckerberg-agrees-to-stay-in-california-despite-billionaires-tax...
1•mikelgan•12m ago•0 comments

US Denaturalizes 2 People Who Gained Citizenship Hiding Criminal Activity

https://www.justice.gov/opa/pr/trump-administration-denaturalizes-two-people-who-gained-us-citize...
2•gnabgib•15m ago•0 comments

Meta's new terms say if its AI makes a mistake, it will be the customer's fault

https://www.cio.com/article/4237395/metas-new-enterprise-contract-says-if-its-ai-makes-a-mistake-...
3•cdrnsf•15m ago•0 comments

Update your Xbox Series S|X Controller through WebUSB

https://dzervas.github.io/xpad-webusb-update/
1•ttouch•19m ago•0 comments

Morph 4K Firmware v4: Simulating the Physics of Retro Video

https://www.pixelfx.co/post/morph-4k-firmware-v4-simulating-the-physics-of-retro-video
1•bpierre•24m ago•0 comments

GameFrame: Agentic Framework for Reverse Engineering Games

https://tanepiper.github.io/game-frame/
3•tanepiper•25m ago•1 comments

Veggie Thursdays and vegan surprises How Ghent became a plant-based paradise

https://www.theguardian.com/food/2026/oct/21/veggie-thursdays-vegan-surprises-ghent-plant-based-p...
1•rdmuser•25m ago•0 comments

Completing a task does not remove the work of managing it

https://civicpivot.blogspot.com/2026/10/the-work-hidden-inside-just-ask.html
1•civicpivot•29m ago•0 comments

They Created a Cell That's Almost Alive. What to Do with It Now?

https://www.nytimes.com/2026/10/14/science/spudcell-adamala-biotic.html
3•CharlesW•30m ago•0 comments

Postgres Recovery_prefetch Considered Harmful and Causing Outages

https://ardentperf.com/2026/10/20/postgres-recovery_prefetch-considered-harmful-and-causing-outages/
1•plaur782•30m ago•0 comments

Qpack amplification attacks against HTTP/3 servers

https://foxio.io/blog/header-accordions-qpack-amplification-attacks-against-http-3-servers
5•kozika•30m ago•0 comments

Dense Sand – Sovereign BYOA AI SoC Platform for AWS

https://aws.amazon.com/marketplace/pp/prodview-iamlatxnobx7o
1•DenseSand•32m ago•0 comments

Make Your Startup More Powerful, Not Just More Profitable

https://www.paulgraham.com/powerful.html
4•mrkn1•34m ago•0 comments

SimulaVR Compute Pack Progress

https://simulavr.com/blog/compute-pack-progress/
1•ai_critic•34m ago•0 comments

A tale of two PHP remote vulnerabilities

https://calif.io/php2shell.x/
1•kozika•35m ago•0 comments

Fil-PHP

https://github.com/pizlonator/fil-c/pull/338
3•SSLy•36m ago•0 comments

Misaligned

https://thezvi.substack.com/p/misaligned
2•7777777phil•37m ago•0 comments

Adversarial interoperability is a team sport

https://pluralistic.net/2026/10/21/brittle-individualism/
1•hn_acker•37m ago•0 comments

EFF's State Legislative Agenda to End ALPR Mass Surveillance

https://www.eff.org/deeplinks/2026/10/effs-state-legislative-agenda-end-alpr-mass-surveillance
2•hn_acker•38m ago•0 comments

Airbnb's new "Reasonable house rules and requests" guidelines for hosts

https://www.airbnb.com/help/article/4168
2•embedding-shape•38m ago•0 comments

39-humanizer, for anyone under 40 who would rather not sound like a bot

https://github.com/OmShiv/39-humanizer
1•OmShankar3•40m ago•0 comments

Show HN: ReviewerAgent and StopContinue, for CC and Codex with Pi

https://github.com/rladr-org/reviewer-agent
1•raylad•40m ago•0 comments

Imaging Black Holes with Julia [video]

https://www.youtube.com/watch?v=eHTuWt3JnZc
1•adgjlsfhk1•41m ago•0 comments

What Lego Lost When It Grew Up

https://www.theatlantic.com/culture/2026/10/lego-adult-market/689051/
1•wltyn•43m ago•2 comments

Oracle Security Alert Advisory – CVE-2026-91403 (eMudhra cert revocation)

https://www.oracle.com/security-alerts/alert-cve-2026-91403.html
19•rayhaanj•46m ago•3 comments

Recompiling the Lion King (SNES, 1994) with an Agent

https://kinduff.com/2026/10/21/recompiling-the-lion-king-with-an-agent/
2•kinduff•46m ago•0 comments

Reactive Optimizers: rewrite the optimizer's source code during training

https://xingjianbai.com/reactive-optimizers/
2•matt_d•46m ago•0 comments

Malleable Systems Collective

https://malleable.systems/
2•so-cal-schemer•46m ago•1 comments
Open in hackernews

Railway Security Disclosure

https://blog.railway.com/p/security-disclosure-wednesday-october-21st
136•passenger•1h ago

Comments

Keloran•53m ago
Well this is certainly interesting, I wonder how their secrets were being stored, I assumed they were stored in vault and then injected with a sidecar, but I guess not
MattIPv4•49m ago
Seen multiple folks in the Railway Discord confirming that their secrets stored in Railway have been used by third parties, so it definitely sounds like the attackers were able to read decrypted environment variables.
MattIPv4•48m ago
Related: https://news.ycombinator.com/item?id=50177413
theturtletalks•42m ago
Postiz founder posted on X that his service is being brought down, his keys were already used and Railway was quiet about this until he spoke up:

https://x.com/wickedguro/status/2113015432974405658

DrawTR•42m ago
Jeez, what a rough past 24 hours for all security folk. Hang in there
jacquesm•40m ago
It's going to be a rough decade.
rectang•29m ago
I really don't know how we are going to deal with accelerated exploit generation via LLMs — it blows up the disclosure processes we've evolved.

The only thing I can think of is hunkering down with static HTML to reduce attack surface, but that's laughably incompatible with many many business requirements.

cortesoft•28m ago
It feels like the only solution is going to be proactively running your own AI agents looking for patches and applying them.
rectang•22m ago
The issue is where to direct the LLM's attention.

Prompting the LLM to "look for vulnerabilities in PHP's multipart upload handling code" is going to find stuff that "look for vulnerabilities in PHP" won't.

paxy•39m ago
It’s funny that every security breach disclosure now has a mandatory “the attacker used AI!” as if it absolves you from having to maintain a secure system. If anything it makes the situation more damning. If it was possible to find exploits using AI then your own red teaming should have found them before the attacker did.
barrrrald•37m ago
This one doesn’t?
arcanemachiner•34m ago
> your own red teaming should have found them before the attacker did

Easier said than done. The defender has to do everything right, all the time, to prevent a headline like this. The attacker only needs to be right once, and it's game over. AI or no AI.

gkoberger•31m ago
Unless they edited the post, they never said this. You should delete this; it's false information.
tokioyoyo•25m ago
Defending is so much harder than attacking. You don't even have to focus on one target, as you can just mass-AI-script a lot of attacks wherever it's possible, and see what sticks. Even if you did everything right, there's always the social-engineering angle, or a single compromised device, or just accidental bad upgrade through supply chain. If you can mass-spread the seed, and plant something that "calls home" constantly, you can just have an agent that would start the attack the minute it gets infected, and go deeper.

It's really a wild west right now, have no idea how we fix it. I'm assuming everything will get more concentrated in big co's hands, as having a very robust security is extremely hard.

Also, we have to agree that younger people will start companies. There will be small companies that cannot afford red teams. There will be companies that are just not equipped, and their profit margins cannot sustain constant attacks.

wateralien•39m ago
Is this our new AI daily reality?
iririririr•28m ago
companies probably have php and Python applications exposed to the web and nobody knows.

the speed of irresponsible deployments is infinite now, by design

esseph•38m ago
Fuck, I thought this was about trains
minraws•38m ago
how are the secrets decrypted, I would like some explanation so that I can have some understanding here, I am starting to feel like all these companies are just meme-ing with their security posture.

Like did someone vibe code too hard, how do env secrets leak for everyone...

9cb14c1ec0•29m ago
Easy, all the bad guys need to do is access whatever system stores the decryption keys.
gbnwl•38m ago
The past two days have been awful as someone who uses Railway.
martinald•36m ago
I have a bad gut feeling that every day is going to look a bit like today with endless urgent patches and breaches.

Fwiw it sounds similar to the hack over the weekend that opencode suffered.

reconnecting•32m ago
I think the only way out is to go back to on-prem for everything, like in the 00s.

LAN cables will go up in price soon.

solyin•28m ago
How does on-prem help if your product is a SaaS.
lawrenceduk•14m ago
You open a SaaS shop where people come to use it
codeulike•10m ago
Instead of being a SaaS it will need to just become a S
80fj0348i•9m ago
You put the SaaS into a Box your customers can install into their datacenter!
yakz•23m ago
It's only just starting. It'll get worse before it gets better.
lijok•36m ago
Damn I guess all those l33t hiring practices were for nothing
rvz•20m ago
Of course, it was a giant scam that tested for nothing and optimized for how to pass a test.

Let's see if companies are ready to hire more people who know what they are doing, against a super-intelligent swarm of millions of agents with cyber capabilities.

alpineman•32m ago
Train wreck
AJRF•32m ago
Do cloud providers have a lock-down mode where you can press 1 button and stop all spend?
eecks•22m ago
Huh, I've been using Railway tonight and this was the first I've seen of this. Just got an email about it now too at 23:36
sergiotapia•22m ago
If you're on Railway do not wait, ROTATE every API key at the source, and do not set new values because Railway is still compromised and not secured.
fbrncci•19m ago
Right just as I take my first day off (for the entire year!) from work to fly out for a vacation. The mother of all security bombs starts dropping all around stacks and services that I am responsible for. How else could have taken time off go for me…
fyzix•18m ago
How is it possible rotate keys if you disable deployments? You're basically telling us to take our services offline...
tossstone•5m ago
I've already taken my applications offline and will be migrating to a new host...
mbreese•21m ago
I didn’t actually see that in this post at all. They mention that the end goal of the attackers might be to get to AI inference providers. But I don’t think they are crying “AI was used” here.

Or did I miss something?

Fordec•22m ago
I fully believe the only sustainable future is that hot patching will need to itself become fully automated. Hot patch now with a minimal allowed amount of line changes to head off the AIs, review it in office hours for auditing and the team is actually awake.

My second belief is that environment variables as a development concept will need to be retired. If it's something that causes a security breach if it by its mere existence on a public git repo, it is information that by design cannot ever pass to a filesystem out of a chain of trust, not by mere development conventions people choose to adhere to.

munificent•19m ago
> hot patching will need to itself become fully automated.

Great, even less control over the things we use and more power ceded to corporations.

Fordec•16m ago
Due to the small nature of hot patching (eg. bump this dependency, rotate these keys, turn off that config, follow the runbook) it will unlikely need to be a frontier model and is likely able to be done LocalLLM/Jev style decision making rather than having to hand all the keys to frontier model operators.
stabbles•8m ago
Hot patching is itself a liability
Sivart13•22m ago
surely once there's been enough patches the code will be perfect though?