Hubert
@hubert3
Followers
1K
Following
6K
Media
252
Statuses
11K
Pentester/Red Teamer. Some code at https://t.co/TIwnZY5vjR
Joined March 2009
đ¨ New blog post: ELEGANTBOUNCER - Catch iOS 0-click exploits without having the samples. Features iOS backup forensics & messaging app scanning for iMessage, WhatsApp, Signal, Telegram & Viber attachments. đ Link ->
8
62
204
NPR EXCLUSIVE: A whistleblower tells Congress and NPR that DOGE may have taken sensitive labor data from the NLRB and hid its tracks. A must-read from @JennaMC_Laugh. https://t.co/9kuJUXaptn
npr.org
A whistleblower tells Congress and NPR that DOGE may have taken sensitive labor data and hid its tracks. "None of that ... information should ever leave the agency," said a former NLRB official.
73
1K
2K
Shingles vax appears to prevent dementia! Our Nature paper is first of many from my group to show this. We use a really neat natural experiment to provide causal (not just correlational!) evidence. Paper: https://t.co/yKWKFhGrUn; video (đ @DrMayaAdam):
34
231
817
There are so many great reasons to be on Signal. Now including the opportunity for the vice president of the United States of America to randomly add you to a group chat for coordination of sensitive military operations. Donât sleep on this opportunityâŚ
335
2K
9K
As OpenAI and Meta introduce LLM-driven searchbots, I'd like to once again remind people that neither LLMs nor chatbots are good technology for information access. A thread, with links: >>
115
1K
5K
Donate an eSIM PLEASE, We are running out and the amount of messages we're getting is CRAZY! The whole team is pressured between the few eSIMs we have and the THOUSANDS sending us. I AM LOSING IT! #ConnectingGaza @Connectinghu_
20
1K
861
The Nobel prize just supercharged corporations' data theft, labor exploitation & plundering of the environment by lending credence to the notion that they have built "superintelligent" things that are gonna either take us out or bring us utopia any day now (who knows which one).
2
17
54
Leading article in NYT this morning: why canât the US do anything to stop Israelâs aggression? Theyâre powerless, bumbling around in the dark looking for their dicks. Oh well 8 paragraphs in: Well, Biden can stop Israel but just doesnât want to? Anyway moving on.
30
487
2K
As @RachelTobac points out, this auto opting in craze has to stop. Just like the EU cookie directive, which yes had a terrible implementation, made many aware of the pervasive nature of ad tracking, we need the same for AI data training
0
5
15
At the height of One Million Checkboxes's popularity I thought I'd been hacked. A few hours later I was tearing up, extraordinarily proud of some brilliant teens. A thread about my favorite story from running OMCB....
I made a website. it's called "one million checkboxes dot com". it has one million checkboxes on it. checking a box checks it for everyone. that's it. have fun!
326
13K
120K
Excellent new essay by Ted Chiang, who gets to the heart of why language and art are inherently about conveying meaning and experience, not just form: https://t.co/agSPvlaLb0
newyorker.com
To create a novel or a painting, an artist makes choices that are fundamentally alien to artificial intelligence.
15
110
457
If only #Crowdstrike consulted @BretHart this all could have been avoided.
oh my god itâs a null ptr deref look at the instruction and r8
1
30
125
Train vendor in Poland is suing Polish ethical hackers who identified logic bombs making trains stop. I'm not sure how the courts will handle this case. But I am sure how it would fare in Court of Justice of European Union. It would be a precedent! Perhaps it's worth it :-)
It's finally happened! NEWAG IP Management just sued us for copyright infringement and unfair competition. Here's a symbolic picture of the lawsuit as a whole: Newag quoting q3k's own code as supposedly their IP :) More: https://t.co/8rS20yv06O
2
30
71
RIP to the surprisingly rich vein of yarns about Aussie politicians liking porn tweets at 3am - and condolences to the specialist AFP team who were asked to investigate them all
Yeah, we are making likes private. Public likes are incentivizing the wrong behavior. For example, many people feel discouraged from liking content that might be "edgy" in fear of retaliation from trolls, or to protect their public image. Soon youâll be able to like without
3
7
75
This is incredibly dangerous. It lays the path for centralized, device-level client side scanning. From detecting 'scams' it's a short step to "detecting patterns commonly associated w/ seeking reproductive care" or "commonly associated w/ providing LGBTQ resources" or
We're testing a new feature that uses Gemini Nano to provide real-time alerts during a call if it detects conversation patterns commonly associated with scams. This protection all happens on-device so your conversation stays private to you. More to come later this year! #GoogleIO
127
2K
6K
Now imagine this AI model doesnât work for you. Imagine it listens to your conversations but it works for and reports to the police. Thatâs where the EU is going.
We're testing a new feature that uses Gemini Nano to provide real-time alerts during a call if it detects conversation patterns commonly associated with scams. This protection all happens on-device so your conversation stays private to you. More to come later this year! #GoogleIO
50
553
2K
Almost like AI hype is NFTs 2.0
This week's newsletter: Less than a year before the Rabbit R1 was announced, Rabbit Inc had another name, and was hyping an NFT metaverse project that it raised $6m to build, promising an MMORPG, a comic, a tv show and a literal rocket shot into space. https://t.co/bdeGdauMcD
0
0
1