max tomago

Coffee and cigarette enjoyer

[max tomago @milloni0]

Are you happy, anon?

[max tomago @milloni0]

@jimhamilton4 Do you have a permit for that AC, mate?

[max tomago @milloni0]

@PabloAntonio Hmm

[max tomago @milloni0]

Collapse of the welfare system

[Jonatan Pallesen @jonatanpallesen]

The demographic future of Europe is looking bleak.

[max tomago @milloni0]

True

[ROY @roy_is_living]

This will get increasingly worse unless fertility rates reverse

In 10-15 years we will likely see the most abnormal dating marketplace in american history

The pool of women 20-29 as a % of total population will keep getting smaller and smaller.

[max tomago @milloni0]

Concering

[Matthew Green @matthew_d_green]

I’m watching how quickly age verification is becoming a part of every policy proposal. None of this is about age, it’s all ultimately about identity.

[max tomago @milloni0]

@ProfesorPingwin Lmaooo

[max tomago @milloni0]

🇵🇱

[Alec Stapp @AlecStapp]

35 years ago, Poland and Iran had similar standards of living.

Now, Poland is about to pass Japan in GDP per capita.

How do we create more Polands?

[max tomago @milloni0]

Proof of humanity

[IT Guy @T3chFalcon]

Google is testing a new reCAPTCHA that asks you to wave your hand at your camera.

The reason: AI can now solve every traditional CAPTCHA. traffic lights, crosswalks, fire hydrants, distorted text. all defeated. bots pass them faster than humans.

So Google's solution is biometrics. your hand. on camera. 21 joint coordinates extracted. analyzed. verified.

Google says: videos deleted after verification. not linked to your identity. audio never recorded.

now here's the timeline of what Google has said about data before:

CAPTCHA was originally designed to help digitize old books and street signs. harmless. useful. then Google used it to label training data for Street View and Google Maps. you were solving puzzles. you were also labeling Google's datasets for free.

And the new system: already bypassed. A user demonstrated it using a virtual camera feeding AI-generated hand animations. the verification designed to prove you're human was defeated by an AI pretending to have a hand.

so to recap:

the old CAPTCHA farmed your attention for training data.
the new CAPTCHA wants your camera and your hand geometry.
it's already broken.
and the bot traffic it's trying to stop just crossed 55% of all internet traffic.

Google needs you to wave at your camera to prove you're human.

The AI watching you wave already knows you are.

It just needed the footage.

[max tomago @milloni0]

@BowTiedPhys I aim for 6-7 hours

[max tomago @milloni0]

Clever attack

[Abdulkadir | Cybersecurity @cyber_razz]

DO NOT touch that keyboard. This is one of the most dangerous attacks circulating right now.

This is called a ClickFix attack. It is not a CAPTCHA. It is not a verification step. It is a social engineering attack designed to make you execute malicious code on your own machine while believing you are proving you are human.

Here is exactly what happens if you follow those steps.

The fake page has already silently copied a malicious PowerShell command to your clipboard without you knowing. It happened the moment the page loaded. You did not click anything. You did not consent to anything. The clipboard was written to in the background by JavaScript running on the page.

When you press Win + R you open the Windows Run dialog. When you press Ctrl + V you paste that malicious command directly into it. When you press Run you execute it with your own permissions on your own machine. No exploit needed. No vulnerability needed. You did it yourself. Willingly. While thinking you were completing a CAPTCHA.

The payload varies. Researchers have documented ClickFix delivering infostealers, remote access trojans, and credential harvesters. The malware executes instantly and silently. By the time the Run dialog closes the damage is done.

The reason this attack works so well is threefold. The fake CAPTCHA looks visually identical to a real one. The instructions sound technical and therefore trustworthy. And critically, you are the one executing the command so endpoint security tools see a legitimate user action rather than an automated attack.

Real CAPTCHAs never ask you to open Run dialogs. Real CAPTCHAs never ask you to paste anything. Real CAPTCHAs never give you keyboard shortcuts.

If a webpage ever asks you to press Win + R for any reason, close the tab immediately.

[max tomago @milloni0]

☕️

[World of Statistics @stats_feed]

☕️ Top coffee consuming countries (kg per capita annually):

🇫🇮 Finland: 9.6
🇳🇴 Norway: 7.2
🇳🇱 Netherlands: 6.7
🇸🇪 Sweden: 6.5
🇸🇮 Slovenia: 6.1
🇦🇹 Austria: 5.5
🇷🇸 Serbia: 5.4
🇩🇰 Denmark: 5.3
🇩🇪 Germany: 5.2
🇧🇪 Belgium: 4.9
🇧🇷 Brazil: 4.8
🇧🇦 Bosnia-Herzegovina: 4.3
🇪🇪 Estonia: 4.2
🇨🇭 Switzerland: 3.9
🇭🇷 Croatia: 3.8
🇩🇴 Dominican Republic: 3.7
🇨🇷 Costa Rica: 3.7
🇲🇰 Macedonia: 3.6
🇮🇹 Italy: 3.4
🇨🇦 Canada: 3.4
🇱🇹 Lithuania: 3.3
🇫🇷 France: 3.2
🇺🇸 USA: 3.1
🇵🇱 Poland: 3.1
🇭🇺 Hungary: 3.1
🇪🇸 Spain: 3
🇰🇷 South Korea: 2.6
🇩🇿 Algeria: 2.6
🇦🇺 Australia: 2.6
🇵🇹 Portugal: 2.6
🇸🇰 Slovakia: 2.5
🇱🇻 Latvia: 2.5
🇬🇷 Greece: 2.4
🇨🇿 Czech Republic: 2.3
🇧🇬 Bulgaria: 2.3
🇹🇳 Tunisia: 2.2
🇸🇦 Saudi Arabia: 2.16
🇸🇬 Singapore: 2.1
🇬🇹 Guatemala: 2
🇷🇴 Romania: 1.9
🇮🇱 Israel: 1.8
🇬🇧 The UK: 1.7
🇹🇭 Thailand: 1.6
🇻🇪 Venezuela:  1.6
🇯🇵 Japan: 1.5
🇨🇴 Colombia: 1.4
🇲🇾 Malaysia: 1.3
🇭🇰 Hong Kong: 1.3
🇳🇿 New Zealand: 1.3
🇬🇪 Georgia: 1.3
🇷🇺 Russia: 1.2
🇲🇽 Mexico: 1.2
🇵🇭 Philippines: 1.2
🇿🇦 South Africa: 0.9
🇦🇷 Argentina: 0.8
🇮🇩 Indonesia: 0.8
🇹🇷 Turkey: 0.4
🇨🇳 China: 0.1
🇮🇳 India: 0.1
🇳🇬 Nigeria: 0.03

🌍 World: 2.0

According to Euromonitor International via Perk Coffee

[max tomago @milloni0]

@KnowledgeArchiv Hmm

[max tomago @milloni0]

If you know, you know

[karel @nextgenkarel]

Karel’s All Together Program is really on Track now

[max tomago @milloni0]

@nextgenkarel Terrible state of things, sad

[max tomago @milloni0]

@memecrashes Hopefully won't happen

[max tomago @milloni0]

If GPT-5.6 is supposed to be a frontier model, OpenAI is now falling behind

[Pankaj Kumar @pankajkumar_dev]

GPT-5.6 Pro Leaks: Coming Thursday

- Knowledge cutoff: December 2025 (previous versions had August 2025).
- Reasoning effort ("Juice Value") increased to 960 (vs 768 on GPT-5.5).
- Vision and image to design replication are much better (it can almost copy designs)
- SVG generation, especially 3D/static SVGs, is excellent and can outperform Fable 5.
- Frontend generation has improved, but Fable/Claude still have the edge.
- Game generation is more stable with fewer visual glitches, though Fable 5 is still better overall.
- Playwright support is in ChatGPT for browser automation.
- GPT-5.6 / GPT-5.6 Pro is most likely coming Thursday.

[max tomago @milloni0]

@robj3d3 Cambodia is the new Bali

[max tomago @milloni0]

The future will be amazing

[International Cyber Digest @IntCyberDigest]

We’ve entered a new era as a species, and this is only the beginning. We’re barely scratching the surface of what AI can do.

[max tomago @milloni0]

Societal collapse

[François Valentin @Valen10Francois]

The surge in welfare spending in the UK in the last 3 years has been absolutely nuts.

Just the nominal increase from 2022/2023 could cover all of its Britain's defence spending (and more)

[max tomago @milloni0]

Big leaps in innovation are a result of questioning assumptions

[tender @tenderizzation]

6-7 decades of computer systems research in a nutshell