frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Show HN: Cerno – CAPTCHA that targets LLM reasoning, not human biology

https://cerno.sh
10•plawlost•1h ago

Comments

opem•1h ago
this game is a rage bait! Try solving on a mobile device. I don't think an elderly person would ever be able to get through this!
king_phil•57m ago
Was impossible for me to solve, too, on mobile.
bestouff•55m ago
Same. I think an AI would have an easier time than me.
ahofmann•56m ago
I managed once to complete the maze, but then it went into "checking" and a spinner forever.
ge96•48m ago
Even when I successfully draw a path from green to red, not touching anything it says "Sorry that's not quite right"
iwalton3•1h ago
Unfortunately it didn't work for me with my trackball mouse. I tried multiple times. This is very likely also problematic for accessibility without another verification pathway, as assistive devices and screen readers would fail the heuristics too.
dcliu•57m ago
Interesting idea but it didn't accept me the first 3-4 times I tried to draw the path on laptop trackpad - going slow led to it failing before I was finished, going fast led to messy paths that were rejected as well. Could something like this work for users with different levels of dexterity?
kccqzy•55m ago
I am on a mobile device and I consistently find that my finger is not precise enough to draw a path; it kept complaining about path “not quite right” where my path just touched the wall a little bit. And when my finger literally obscures the screen while drawing, it essentially means I have to memorize an entire path before beginning.

This requires significant spatial thinking skills and short-term memory for a human and it is absolutely unfair for a human to expend that much cognitive power to access a website.

happytoexplain•51m ago
Everybody is saying it's impossible on mobile/trackpads, but I couldn't do it with a mouse either :(
CoUHKT57BSe•47m ago
I'm assuming this is an april fools joke... cause it just don't work at all as a CAPTCHA.
ranger_danger•39m ago
The description doesn't make sense either... like "Color-word interference at maze decision points", but there are no colors (or words) at the decision points.
ravenical•47m ago
Is the idea that if you /pass/, you're an AI?
_diyar•45m ago
I‘m as confused as the other commenters, especially since it‘s unusable on an iPad.
simlevesque•44m ago
I did it correctly with a mouse and was rejected. Then I drew a straight line and was rejected. Then I closed the tab.
Surac•38m ago
Does not work on a touchscreen. Can’t see maze under finger and smh it always kicks me in the middle of the drawing
ceayo•35m ago
am i a robot or is this just way too hard?
1attice•15m ago
Did not work for me (kept getting "something's not right") but I applaud the concept.

I worry a bit about accessibility but that is a problem all CAPTCHAs have

DHolzer•14m ago
like the idea, but i cant see what i'm doing when my finger coveres the screen

Stackra – free website audit with platform-aware scoring and AI readiness check

https://stackra.app/
1•LBeck84•22s ago•0 comments

Income percent for all food, food at home and food not at home, 1960-2019 (2020)

https://ers.usda.gov/data-products/charts-of-note/chart-detail?chartId=100002
1•NaOH•2m ago•0 comments

Craft Is Untouchable

https://www.chrbutler.com/craft-is-untouchable
3•delaugust•2m ago•0 comments

AWS Service Availability Updates

https://aws.amazon.com/about-aws/whats-new/2026/03/aws-service-availability/
1•alexbilbie•2m ago•0 comments

Ask HN: Linuxdoors.com What should be on there?

1•vibecoderowl•3m ago•0 comments

OpenAI raises $122B to accelerate the next phase of AI

https://openai.com/index/accelerating-the-next-phase-ai/
4•alvis•5m ago•0 comments

Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs

https://prismml.com/
1•PrismML•6m ago•0 comments

Security releases of JupyterHub, Auth0, LTI authenticators

https://discourse.jupyter.org/t/ann-security-releases-of-jupyterhub-auth0-lti-authenticators/38480
1•julienchastang•7m ago•0 comments

An effect notation based on with-clauses and blocks

https://blog.yoshuawuyts.com/a-with-based-effect-notation
2•lukastyrychtr•7m ago•0 comments

The Decline of the West [video]

https://www.youtube.com/watch?v=YwtVOgHJU7I
1•joe_mamba•8m ago•0 comments

QuantumForge – Browser-based quantum chemistry simulations on IBM Quantum

https://quantumforgelabs.io
1•isurfclouds•8m ago•1 comments

Cost Effective SOTA Vulnerability Detection with Open Source RL

https://depthfirst.com/post/dfs-mini1-agent
1•ejcho•9m ago•0 comments

Kaggle launches auto exams for zero-setup agent evals of prompt injection, etc.

https://www.kaggle.com/blog/standardized-agent-exams
1•art82135•11m ago•0 comments

The Nuclear Safety Protections in Federal Crosshairs

https://undark.org/2026/03/31/federal-nuclear-safety-protections/
1•EA-3167•11m ago•0 comments

I published the architecture Anthropic is now shippingstamped proof

https://nicholasjbogaert.substack.com/p/i-published-the-architecture-anthropic
3•nicbogaert•12m ago•1 comments

TMZ Has a New Obsession: Vacationing Members of Congress

https://www.notus.org/congress/tmz-congress-coverage-dhs-shutdown-tabloid-paparazzi
2•OrangePilled•13m ago•0 comments

Functional hierarchy of the human neocortex across the lifespan

https://www.nature.com/articles/s41586-026-10219-x
1•bookofjoe•14m ago•0 comments

Redesign any website with four AI models side by side

https://shuffle.dev/ai-website-redesign
3•avoc777•15m ago•0 comments

Cisco source code stolen in Trivy-linked dev environment breach

https://www.bleepingcomputer.com/news/security/cisco-source-code-stolen-in-trivy-linked-dev-envir...
5•_____k•17m ago•0 comments

Scaling Laws: The Lawfare Playlist on AI [video]

https://www.youtube.com/playlist?list=PL9f-8IUHQF3lNz1hhIIU1AyQM9M6hmP4m
1•verdverm•17m ago•0 comments

Ministack (Replacement for LocalStack)

https://ministack.org/
4•kerblang•19m ago•0 comments

APS: Open specification for AI agent policies

https://agentpolicyspecification.github.io
1•pascalwilbrink•21m ago•0 comments

Show HN: AI writes the code, but who answers in the interview?

https://sharpskill.dev/en
1•GiornoJojo•21m ago•0 comments

Amazon pulled Kindle update 5.19.3 a few days after release

https://old.reddit.com/r/kindle/comments/1s7lo50/5193_officially_back_to_5192_on_kindle_website/
2•seam_carver•23m ago•0 comments

OpenAI Valued at $852B After Completing $122B Round

https://www.bloomberg.com/news/articles/2026-03-31/openai-valued-at-852-billion-after-completing-...
3•matthieu_bl•23m ago•0 comments

Top Five (Wrong) Reasons You Don't Have Testers (2000)

https://www.joelonsoftware.com/2000/04/30/top-five-wrong-reasons-you-dont-have-testers/
3•Austin_Conlon•23m ago•0 comments

Codex Plugin for Claude Code

https://community.openai.com/t/introducing-codex-plugin-for-claude-code/1378186
1•grodriguez100•25m ago•0 comments

Show HN: Fixing Claude Code's amnesia with persistent memory

https://github.com/agynio/claude-map-reduce-memory
1•NBenkovich•29m ago•2 comments

OpenHarness – Composable TypeScript SDK for building powerful agent harnesses

https://www.open-harness.dev
2•MaxGfeller•29m ago•1 comments

How I Fixed My Headaches

https://chillphysicsenjoyer.substack.com/p/how-i-fixed-my-headaches
4•sebg•29m ago•0 comments