Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Vercel April 2026 security incident

bleepingcomputer.com

Post reply on HN
Print view View original

371–380 of 540 posts

Re: Vercel April 2026 security incident

#371

Post by 00deadbeef »

cleaning wrote: post #279
0xmattf wrote: post #82

Earlier quoted context omitted.

> The only reason to dramatically overpay for the hosting resources they provide is because you expect them to expertly manage security and stability. This and because it's so convenient to click some buttons and have your application running. I've stopped being lazy, though. Moved everything from Render to linode. I was paying render $50+/month. Now I'm paying $3-5. I would never use one of those hosting providers a…

If you're only paying $3-5 on Linode then your level of usage would probably be comfortably at $0 on Vercel.

What if they have an actual back-end with long-running processes and scheduled tasks?

Re: Vercel April 2026 security incident

#372

Post by UltraSane »

ryanscio wrote: post #295
nettlin wrote: post #180

They just added more details: > Indicators of compromise (IOCs) > Our investigation has revealed that the incident originated from a third-party AI tool whose Google Workspace OAuth app was the subject of a broader compromise, potentially affecting hundreds of its users across many organizations. > We are publishing the following IOC to support the wider community in the investigation and vetting of potential malicio…

https://x.com/rauchg/status/2045995362499076169 > A Vercel employee got compromised via the breach of an AI platform customer called http://Context.ai that he was using. > Through a series of maneuvers that escalated from our colleague’s compromised Vercel Google Workspace account, the attacker got further access to Vercel environments. > We do have a capability however to designate environment variables as “non-sens…

Production network control plane must be completely isolated from the internet with a separate computer for each. The design I like best is admins have dedicated admin workstations that only ever connect to the admin network, corporate workstations, and you only ever connect to the internet from ephemeral VMs connected via RDP or similar protocol.

Re: Vercel April 2026 security incident

#374

Post by leduyquang753 »

lmm wrote: post #320
nikcub wrote: post #116

Claude Code defaulting to a certain set of recommended providers[0] and frameworks is making the web more homogenous and that lack of diversity is increasing the blast radius of incidents [0] https://amplifying.ai/research/claude-code-picks/report

Is that bad? I would think having everyone on the same handful of platforms should make securing them easier (and means those platforms have more budget to to so), and with fewer but bigger incidents there's a safety-of-the-herd aspect - you're unlikely to be the juiciest target on Vercel during the vulnerability window, whereas if the world is scattered across dozens or hundreds of providers that's less so.

When everyone uses the same handful of platforms, then everyone becomes the indirect target and victim of those big incidents. The recent AWS and Cloudflare outages are vivid examples. And then the owners of those platforms target everyone with their enshittification as well to milk more and more money.

Re: Vercel April 2026 security incident

#375

Post by uecker »

eddythompson80 wrote: post #263
slopinthebag wrote: post #251

Earlier quoted context omitted.

Yeah definitely no difference between GNU coreutils and some vibe coded AI tool released last month that wants full oAuth permissions.

I’m not joking, but weirdly enough, that’s what most AI arguments boil down to. Show me what the difference is while I pull up the endless CVE list of which ever coreutils package you had in mind. It’s a frustrating argument because you know that authors of coreutils-like packages had intentionality in their work, while an LLM has no such thing. Yet at the end, security vulnerabilities are abundant in both. The AI ma…

"endless list of CVE" seems rather exaggerated for coreutils. There are only very few CVEs in the last decade and most seem rather harmless.

Re: Vercel April 2026 security incident

#376

Post by arch-choot »

cleaning wrote: post #279
0xmattf wrote: post #82

Earlier quoted context omitted.

> The only reason to dramatically overpay for the hosting resources they provide is because you expect them to expertly manage security and stability. This and because it's so convenient to click some buttons and have your application running. I've stopped being lazy, though. Moved everything from Render to linode. I was paying render $50+/month. Now I'm paying $3-5. I would never use one of those hosting providers a…

If you're only paying $3-5 on Linode then your level of usage would probably be comfortably at $0 on Vercel.

Repeating a prior comment I've made about this[0]: I run a rust webserver on a €4 VPS from hetzner that serves 300M (million) requests a day.

From what I can figure out, Vercel charges "$0.60 per million invocations" [1], which would cost me $180 per day.

[0] https://news.ycombinator.com/item?id=47611454 [1] https://vercel.com/docs/functions/usage-and-pricing#invocati...

Re: Vercel April 2026 security incident

#377

Post by uecker »

pianopatrick wrote: post #341
lijok wrote: post #221

Earlier quoted context omitted.

This isn't a web development concept. It's the unix philosophy of "write programs that do one thing and do it well" and interconnect them, being taken to the extremes that were never intended. We need a different hosting model.

Just throwing it out there - the Unix way to write software is often revered. But ideas about how to write software that came from the 1970s at Bell Labs might not be the best ideas for writing software for the modern web. Instead of "programs that do one thing and do it well", "write programs which are designed to be used together" and "write programs to handle text streams", I might go with a foundational philosoph…

I do not see what this has to do with Unix. The problem is not that programs interoperate or handle text streams, the problem is a) the supply chain issues in modern web-software (and thanks to Rust now system-level) development and b) that web applications do not run under user permissions but work for the user using token-based authentication schemes.

Re: Vercel April 2026 security incident

#379

Post by Foobar8568 »

nnurmanov wrote: post #353
Vates wrote: post #339

When one OAuth token can compromise dev tools, CI pipeline, secrets and deployment simultaneously, something architectural has gone wrong. Vercel have had React2Shell (CVSS 10), the middleware bypass (CVSS 9.1), and now this, all within 12 months. At what point do we start asking questions about the concentration of trust in the web ecosystem? It's funny that at the engineering level we are continuously grilled in in…

You have no idea how indifferent security officers can be-even when you point out critical issues. The other day, we flagged that a customer’s database had users with excessive privileges. Their only question: “Can this be exploited from the outside?” No, but most breaches today come from compromised internal accounts that are then used to break everything.

What's the problem to have local API connected in HTTP? We are within the enterprise network.

And that's how I passed for a annoying "PM". With half of the program management complaining that I was slowing down things until 6m later, the head of risk management told them to get lost.

Re: Vercel April 2026 security incident

#380

Post by jongjong »

nikcub wrote: post #116

Claude Code defaulting to a certain set of recommended providers[0] and frameworks is making the web more homogenous and that lack of diversity is increasing the blast radius of incidents [0] https://amplifying.ai/research/claude-code-picks/report

Yes, this is a genuine problem with AI platforms. It does sometimes feel like they're suspiciously over-promoting certain solutions; to the point that it's not in the AI platform's interest.

I know what it's like being on the opposite side of this as I maintain an open source project which I started almost 15 years ago and has over 6k GitHub stars. It's been thoroughly tested and battle-tested over long periods of time at scale with a variety of projects; but even if I try to use exact sentences from the website documentation in my AI prompt (e.g. Claude), my project will not surface! I have to mention my project directly by name and then it starts praising it and its architecture saying that it meets all the specific requirements I had mentioned earlier. Then I ask the AI why it didn't mention my project before if it's such a good fit. Then it hints at number of mentions in its training data.

It's weird that clearly the LLM knows a LOT about my project and yet it never recommends it even when I design the question intentionally in such a way that it is the perfect fit.

I feel like some companies have been paying people to upvote/like certain answers in AI-responses with the intent that those upvotes/likes would lead to inclusion in the training set for the next cutting-edge model.

It's a hard problem to solve. I hope Anthropic finds a solution because they have a great product and it would be a shame for it to devolve into a free advertising tool for select few tech platforms. Their users (myself included) pay them good money and so they have no reason to pander to vested interests other than their own and that of their customers.

Post reply on HN