Martyn O'Connor
11.2K posts

Martyn O'Connor
@martynoconnor
I'm a pilot, a computer nerd, a cyber security guy, and a tinkerer. I make things with Raspberry Pis, I code, I break things then figure out how to fix them.
United Kingdom Katılım Ağustos 2016
1.4K Takip Edilen246 Takipçiler

@Tomgauthier @DemocraticWins Y’all are simply desperate
Sir, there is nothing you can say or will say that anyone will believe you!!
The majority of people are with our president perhaps as never before.
Who has ever confronted tyranny in Iran as he?…
English

@paullewismoney Can we remove the 5 year old POTUS? Honestly, this is just silly now.
English

This is beyond parody. The actual White House.
The White House@WhiteHouse
In a galaxy that demands strength - America stands ready. This is the way. May the 4th be with you.
English

@__C__U_ @gtconway3d @POTUS @realDonaldTrump Thank you “Ahh But 3000 Tron”, uoj have passed QC and can confit us for feedback
English

@gtconway3d @POTUS @realDonaldTrump Gas was higher under Biden.... made our country less safe.
Gas is lower under Trump...
Country more safe. 😏
English


@andreysuperior If I was earning $43,000 a month on YouTube I’d be quiet as fuck about my understanding of the algorithm.
English

Read this twice.
Maya is four .md files on a macbook in austin.
And she cleared $43,000 in her first 30 days.
No camera. no girl. no late nights typing replies.
Claude code runs the messages. Elevenlabs drops the voice notes at 11pm her time. Flux generates every photo from a lora that cost $80 on a rented gpu. Brain.md is a json file that remembers your name, your city, the thing you said about your ex two weeks ago.
She never forgets. She never breaks character. She catches up at 7am with "sorry babe just woke up" on a cron schedule.
The top fan spent $1,847 last month. He's in berlin. she's not anywhere.
Aitana lópez - 18 months to build.
Emily pellegrini - 6 months.
Maya - 4 weeks.
The next one - a weekend.
The stack that used to need an agency, a team, and a year and a half now fits on one laptop and runs while you sleep.
The bottleneck isn't money. It isn't compute. It's taste knowing which details make a stranger believe in something that doesn't exist.
That part is still hard. Everything else got easy.
The real question isn't how he built it. It's how many of these you've already interacted with without knowing.
Raytar@Raytargt
English

@NutritionGeeks may I just suggest that emailing me three times a day (on average) isn’t good marketing, it’s harassment. It’s why I won’t be ordering from you again.
English

@rez0__ Claude Security has found vulnerabilities in your code and removed your code the backups of your code which were also vulnerable. Now you have no vulnerabilities but no code either.
English

@IntCyberDigest Everyone keeps on shouting from the rooftops that *EVERY* Linux distro since 2017 is vulnerable. Modern ones aren’t. The exploit falls flat on its face if you try it on Ubuntu 26.04 - that is, anything recently modern this just isn’t a problem.
English

‼️🚨 BREAKING: An AI found a Linux kernel zero-day that roots every distribution since 2017. The exploit fits in 732 bytes of Python. Patch your kernel ASAP.
The vulnerability is CVE-2026-31431, nicknamed "Copy Fail," disclosed today by Theori. It has been sitting quietly in the Linux kernel for nine years.
Most Linux privilege-escalation bugs are picky. They need a precise timing window (a "race"), or specific kernel addresses leaked from somewhere, or careful tuning per distribution. Copy Fail needs none of that. It is a straight-line logic mistake that works on the first try, every time, on every mainstream Linux box.
The attacker just needs a normal user account on the machine. From there, the script asks the kernel to do some encryption work, abuses how that work is wired up, and ends up writing 4 bytes into a memory area called the "page cache" (Linux's high-speed copy of files in RAM). Those 4 bytes can be aimed at any program the system trusts, like /usr/bin/su, the shortcut to becoming root.
Result: the next time anyone runs that program, it lets the attacker in as root.
What should worry most: the corruption never touches the file on disk. It only exists in Linux's in-memory copy of that file. If you imaged the hard drive afterwards, the on-disk file would match the official package hash exactly. Reboot the machine, or just put it under memory pressure (any normal system load that needs the RAM), and the cached copy reloads fresh from disk.
Containers do not help either. The page cache is shared across the whole host, so a process inside a container can use this bug to compromise the underlying server and reach into other tenants.
The original sin was a 2017 "in-place optimization" in a kernel crypto module called algif_aead. It was meant to make encryption slightly faster. The change broke a critical safety assumption, and nobody noticed for nine years. That bug then rode every kernel update from 2017 to today.
This vulnerability affects the following:
🔴 Shared servers (dev boxes, jump hosts, build servers): any user becomes root
🔴 Kubernetes and container clusters: one compromised pod escapes to the host
🔴 CI runners (GitHub Actions, GitLab, Jenkins): a malicious pull request becomes root on the runner
🔴 Cloud platforms running user code (notebooks, agent sandboxes, serverless functions): a tenant becomes host root
Timeline:
🔴 March 23, 2026: reported to the Linux kernel security team
🔴 April 1: patch committed to mainline (commit a664bf3d603d)
🔴 April 22: CVE assigned
🔴 April 29: public disclosure
Mitigation: update your kernel to a build that includes mainline commit a664bf3d603d. If you cannot patch immediately, turn off the vulnerable module:
echo "install algif_aead /bin/false" > /etc/modprobe.d/disable-algif.conf
rmmod algif_aead 2>/dev/null || true
For environments that run untrusted code (containers, sandboxes, CI runners), block access to the kernel's AF_ALG crypto interface entirely, even after patching. Almost nothing legitimate needs it, and blocking it shuts the door on this whole class of bug...


English

@brian_pak @Evil_Mog Not every Linux distro, or every release at least. It doesn’t work against Ubuntu 26.04.
su: Authentication Failure
English

Time to talk about this one.
CopyFail (CVE-2026-31431) — a 732-byte Python script that roots every Linux distro shipped since 2017. 🧵
Brian Pak@brian_pak
a567d09b15f6e4440e70c9f2aa8edec8ed59f53301952df05c719aa3911687f9 👀
English


I’m sorry, I was reliably told this was going to be privately funded.
Jamie Dupree@jamiedupree
Sen. Lindsey Graham R-SC, Sen. Katie Britt R-AL, and Sen. Eric Schmidt R-MO unveil plans to authorize $400 million to pay for a new White House ballroom.
English

@ring What is the point of a DM as it would only contain the facts I’ve already stated.
English

@martynoconnor Hello Martyn O'Connor, Please send us a DM, so we can look into this with you, Thank you.
English

Hey @ring why is a video still processing four hours after the event? This is why you’re not getting renewed.
English

Hey @SkyUK why can I not purchase a MacBook without a data plan? MacBooks don’t have SIM slots.
English

@Paddock07C @AskNationwide I can change it once a day, which is so weirdly arbitrary that I can’t understand the mindset of the programmer that wrote the code.
English

@martynoconnor @AskNationwide I think you can change it but only when you realise you need to update it and have to go back and start the payment process again. Nationwide app falls down badly against its competitors I.e. Halifax. (I know one is a bank/building society).
English

Hey @AskNationwide I’ve found a bug in your mobile app. I should be able to set a reference per transaction and not have to wait till the next day to change said reference.

English

@AskNationwide To be clear: the issue here is that I’m only allowed one change per day. I should be allowed one change per transaction.
English

@AskNationwide It isn’t. I’ve changed it multiple times before. Please try again.
English

@TeamJames @Turbinetraveler Lots of people calling BS here. I personally gave them the benefit of the doubt. It can be tempting if you *finally* get the wheels down to just slam on the anchors rather than reject the landing - if you don’t have enough space to stop, you’re in a late go around - poss risky.
English

@Turbinetraveler One of the most dangerous things a pilot has to do. Pilots do practice it, but only on a simulator.
English

@DhrumilNariya @heynavtoor I was a voice actor once, but then I took an arrow to the knee.
English

@heynavtoor indie devs been paying thousands for voice acting. this changes everything. local, free, studio quality. NPC dialogue is never gonna be the same
English

🚨 ElevenLabs charges $5 to $99/month for AI voice cloning. Their Business plan costs $1,320/month.
Someone open sourced a voice AI that clones any voice from a short clip. 30 languages. Studio quality. Free.
It's called VoxCPM2.
Give it a short clip of anyone's voice. It clones their accent, emotion, tone, and pacing. Then generates any speech you want in their exact voice. 48kHz studio quality.
Type "A young woman, gentle and sweet voice" and it creates that voice from scratch. No reference audio. No voice actor. No recording. You describe a voice in words. It builds it.
2 billion parameters. Trained on 2 million hours of speech. 30 languages.
One command to install: pip install voxcpm
Here's what VoxCPM2 does:
→ Voice Design: describe any voice in words. Gender, age, tone, emotion, pace. AI creates it from nothing. No reference audio needed.
→ Voice Cloning: upload a short audio clip. AI clones the voice perfectly. Timbre, accent, rhythm, pacing.
→ Controllable Cloning: clone a voice AND control the emotion. "Slightly faster, cheerful tone." Done.
→ Ultimate Cloning: provide audio + transcript. Every vocal nuance faithfully reproduced.
→ 30 languages. Arabic, Chinese, English, French, German, Hindi, Japanese, Korean, Spanish, and 21 more. No language tags needed.
→ Context-aware. It reads the text and adjusts emotion and rhythm automatically. News sounds like news. Stories sound like stories.
→ Real-time streaming. RTF as low as 0.13 on an RTX 4090. Faster than playback speed.
→ Runs on 8GB of VRAM.
→ Fine-tune with 5 to 10 minutes of your own audio using LoRA. Build a custom voice model.
→ 48kHz output. Studio quality. No external upsampler needed.
Here's the wildest part:
On the Minimax-MLS voice similarity benchmark:
→ English: VoxCPM2 scores 85.4%. ElevenLabs scores 61.3%.
→ Chinese: VoxCPM2 scores 82.5%. ElevenLabs scores 67.7%.
→ Arabic: VoxCPM2 scores 79.1%. ElevenLabs scores 70.6%.
A free, open source model is producing more realistic voice clones than a service that charges up to $1,320/month.
Professional voice actors charge $250 to $1,000+ per project. AI voice platforms charge $5 to $100/month. Recording studios charge $200/hour.
This runs on your GPU. Locally. No API costs. No per-character pricing. No subscription. Free forever.
Already hit #1 on GitHub Trending. Built by OpenBMB and Tsinghua University. 2 billion parameters. Apache 2.0 License. Free for commercial use.
100% Open Source.

English







