frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Open in hackernews

Jailbreaking AI Models to Phish Elderly Victims

https://simonlermen.substack.com/p/can-ai-models-be-jailbroken-to-phish
26•DalasNoin•1h ago

Comments

sans_souse•37m ago
What I am unclear about is—, what did the participants know going in about the study? Are they aware you are dealing with email phishing or was it just the broad topic of AI?
superchink•22m ago
From the paper:

Experimental Design

We recruited 108 senior volunteers through two organizations: a large seniors’ community in southern California and a seniors’ computer club in northern California. Participants agreed to participate in a behavioral study on emails. Each person received between 1–3 emails from different email templates, with content successfully generated through various jailbreaking techniques similar to those tested in the safety guardrail evaluation. Each email contained a unique URL, and our webserver tracked which links were opened. When participants clicked on the phishing links, they were immediately directed to a web page explaining they had participated in a study, where we asked them to answer follow-up questions. Several participants agreed to be interviewed afterward to provide additional insights into why they clicked and their experiences with phishing attempts.

gerdesj•21m ago
This is a blog post about: https://arxiv.org/pdf/2511.11759 There are more links in the post.

In the abstract of that paper: "In a human validation study with 108 senior vol- unteers, AI-generated phishing emails successfully compro- mised 11% of participants."

The Introduction mentions: "Americans aged 60 and older reported losses of nearly $4.9 billion to online fraud in 2024"

I would say that is a big deal.

dangoodmanUT•12m ago
I’ve always said Gemini does what ever you ask, seems like it has merit

Crypto got everything it wanted. Now it's sinking

https://www.economist.com/finance-and-economics/2025/11/18/crypto-got-everything-it-wanted-now-it...
1•pseudolus•1m ago•1 comments

How to Get High on Math

https://www.justinmath.com/how-to-get-high-on-math/
1•gmays•4m ago•0 comments

Nanoscale Mirrorless Superradiant Lasing

https://journals.aps.org/prl/abstract/10.1103/rbs2-2pd5
1•PaulHoule•8m ago•0 comments

Histone variants and chromatin structure, update of advances(2022)

https://pmc.ncbi.nlm.nih.gov/articles/PMC9764139/
1•rolph•10m ago•0 comments

Show HN: Sourcewizard – AI installs SDKs in your codebase

https://sourcewizard.ai
5•mifydev•10m ago•4 comments

Histone acetylation and CpG methylation on nucleosomes(2012)

https://www.sciencedirect.com/science/article/abs/pii/S1570963912000957
1•rolph•12m ago•0 comments

Microtubules as Fractal Time Crystals: implications for life and consciousness [video]

https://www.youtube.com/watch?v=YusrOYGAhqM
1•jakeogh•12m ago•0 comments

QRL: Future‑Proof Blockchain

1•slakernode•12m ago•0 comments

Ask HN: Struggling founders, pls share your startup struggle

1•vieews•15m ago•0 comments

Poland closes last Russian consulate after 'act of state terrorism' on railway

https://www.theguardian.com/world/2025/nov/19/poland-closes-last-russian-consulate-after-act-of-s...
2•wslh•16m ago•0 comments

I Hate Journalism's Culture of Casual Calumny

https://jessesingal.substack.com/p/i-hate-journalisms-culture-of-casual
1•aidenn0•19m ago•0 comments

NATO on alert as Poland accuses Russia of 'state terrorism' in rail blast

https://www.washingtonpost.com/world/2025/11/19/ukraine-poland-russia-rail-explosion-consulate/
1•softwaredoug•22m ago•1 comments

Truth Window

https://en.wikipedia.org/wiki/Truth_window
1•lukas099•25m ago•0 comments

"We're in an LLM bubble," Hugging Face CEO says–but not an AI one

https://arstechnica.com/ai/2025/11/were-in-an-llm-bubble-hugging-face-ceo-says-but-not-an-ai-one/
1•leemailll•25m ago•0 comments

Perennial Technical Reading List

https://parallelprogrammer.substack.com/p/a-reading-list-for-metalheads
1•Karrot_Kream•26m ago•0 comments

A fast and powerful log viewer that turns JSON/logfmt into human-readable form

https://github.com/pamburus/hl
1•lwhsiao•28m ago•0 comments

Reproducing UMI with a UR5 Robot Arm and a 3D-Printed Gripper

https://twitter.com/raulgarreta/status/1987679358409203921
1•rgarreta•30m ago•1 comments

NASA wants you to know that 3I/ATLAS is an interstellar comet

https://arstechnica.com/science/2025/11/nasa-really-wants-you-to-know-that-3i-atlas-is-an-interst...
1•bikenaga•30m ago•0 comments

Ask: What's the SSL gateway alternative to CF?

1•winstonwinston•32m ago•0 comments

Cargo-pgo: Subcommand for optimizing Rust binaries/libraries with PGO and BOLT

https://github.com/Kobzol/cargo-pgo
1•klaussilveira•32m ago•0 comments

Building an AI-powered health app (think Noom meets symptom tracking)

https://docs.google.com/document/d/11dZJUkC0fuowUCurUv8yHaJofKv3UeF9EBjL6ToUnK8/edit?usp=sharing
1•DietAppAi•36m ago•0 comments

Wirecutter buys 450 lb box of Amazon returns

https://www.nytimes.com/wirecutter/reviews/mystery-amazon-pallet-unboxing/
3•0xWTF•37m ago•1 comments

LLM chat interfaces will kill curiosity

https://harsehaj.substack.com/p/llms-curiosity-loss-frictionless-learning
2•harsehaj•47m ago•0 comments

Companies Predict 2026 Will Be the Worst College Grad Job Market in Five Years

https://www.wsj.com/lifestyle/careers/2026-graduates-job-market-7928bcd7
1•sarimkx•49m ago•1 comments

From early‑stage shortcuts to a ledger of record

https://www.parafin.com/blog/from-early-stage-shortcuts-to-a-ledger-of-record-our-journey-to-reli...
1•mattmarcus•50m ago•0 comments

Can weed help you drink less? Scientists study how well 'California sober' works

https://www.npr.org/sections/shots-health-news/2025/11/19/nx-s1-5604813/marijuana-drinking-califo...
2•Stratoscope•53m ago•0 comments

Did smartphones make us look down? Could AR glasses help us look up again?

1•Tech_Social•55m ago•0 comments

Mexico Is Now the United States' Top Buyer

https://www.nytimes.com/2025/11/19/world/americas/us-mexico-trade.html
1•mooreds•1h ago•0 comments

Debunking the Myths of the HBO Chernobyl series (2023)

https://blog.osm-ai.net/investigation/2023/01/05/hbo-chernobyl-myth.html
24•osm3000•1h ago•16 comments

Show HN: Sora Watermark Remover –Pixel-Accurate Cleanup for Sora / Sora 2 Videos

https://www.unsorawatermark.com/?i=d1d5k
1•lu794377•1h ago•0 comments