mkbhru

don't die poor | human

[mkbhru @mkbhru]

world still looks alive to me
https://t.co/olPIjeAYDK
finally read it in June 2026. late by a whole year.
Plot twist: it’s even scarier now because half the predictions already happened 😂
My slightly panicked notes below 🧵👇 #AI #AGI #AI2027

[Rahul 🥷 @themishra4402]

4GB RAM on Linux in 2026:

[さらさら @sarasara_aiart]

長押しで変化？ 𓂃𓈒໒꒱ 𓏸*˚

[mkbhru @mkbhru]

what you know yo what you are... Aye..

[Qadeer Ahmad Shaikh @qdongly]

@Google The interesting shift isn't selfie login itself. It's that security is moving from what you know to who you are. The future won't be stronger passwords,it'll be making passwords disappear altogether.

[GitHub Projects Community @GithubProjects]

Rclone is a command-line program that syncs files and directories across 40+ cloud storage providers, functioning like rsync for cloud services.

- Supports 40+ storage providers including S3, Google Drive, Dropbox, and Backblaze B2
- Encrypts data before uploading to cloud storage
- Preserves timestamps and verifies checksums during sync
- Mounts remote storage as a local filesystem via FUSE

[mkbhru @mkbhru]

@wasmerio FOXVERSE

[Wasmer @wasmerio]

Firefox compiled to WebAssembly, running inside of your browser 🤯

[mkbhru @mkbhru]

@Seraphinaf5 71

[mkbhru @mkbhru]

You aren't limited by your resources, only your resourcefulness!
#mtmlsfl

[mkbhru @mkbhru]

If you don't know what to do - build potential!
#mtmlsfl

[0xSero @0xSero]

Not your weights not your GF

[mkbhru @mkbhru]

@KChaiWrites @Pallavi_345 Cmux

[mkbhru @mkbhru]

@Anaya_sharma876 mine is distrohoppin🍾

[mkbhru @mkbhru]

@kunoo Ranching☠️

[The Rundown AI @TheRundownAI]

Meta got a brain-to-text decoder to 61% word accuracy, reading raw signals from outside the skull without any implants or surgery.

The previous best for reading the brain without surgery = ~8%.

It learned from 9 volunteers, who each sat 10 hours inside a brain scanner and typed while the system read along.

One AI model read the raw brain signals as they typed, and a language model filled in the meaning.

The top volunteer hit 78%, with over half of their sentences came back with one word wrong at most.

This is v2 of Brain2Qwerty. v1 was published today in Nature, and only decoded one character at a time.

v2 jumped to whole words and their meaning, with Meta releasing the training code for both.

And accuracy keeps climbing with more data, meaning the gap with surgical implants might close on scale alone.

[mkbhru @mkbhru]

@Palak3312 Deadlock🔒

[mkbhru @mkbhru]

@Its_lakshya_ai English ✨

[mkbhru @mkbhru]

@hoshmandEnt @payaldixit12323 Maintenance mode☠️