Sóc poliglota, parlo català, valencià i mallorquí. Membre de @softcatala i marejant la perdiu a @cohere_labs Coses que faig 🔨: https://xitter.cf/t.co/iHjgwaGqVB
Guinardó, Catalonia
Joined February 2010
- Tweets19.2K
- Following1.9K
- Followers4.2K
- Likes17.1K
Jordi Mas retweeted
Seny, la IA catalana que vigila els agents empreses.ara.cat/innovacio/s…
AWS ja té reservada molta de la seva capacitat computacional al núvol per a 2027, en gran part per la demanda dels grans laboratoris d'IA.
Qualsevol que treballi al núvol, ha patit allò de què no hi ha màquines amb GPUs modernes i has d'anar a altres regions a veure si tens sort
Five years into his run as CEO, Andy Jassy has nearly tripled the size of AWS, grown Amazon's revenue by about 80%, and expects to spend $220 billion this year to win AI. Amazon's AI business hit a $25 billion run rate in three years, growing several hundred times faster than early AWS.
Jassy joins a16z's Ben Horowitz and Erik Torenberg to discuss the decisions behind that run, from rebuilding Amazon's fulfillment network after the pandemic to the AWS origin story.
They also cover why data centers became a political fight, Amazon's custom chips, its stakes in Anthropic and OpenAI, why inference needs to get 10x cheaper, AI's most underrated risk, and why every company will have to reinvent its products from scratch, plus Jassy's advice for anyone who's 18 today.
0:47 How Amazon has changed since 2021
1:51 How the pandemic reshaped retail
3:30 Why AI supercharged AWS's growth
5:20 When leadership principles clash
9:36 $60M vs. $25B: early AWS vs. early AI
13:20 How the AI CapEx cash cycle works
15:44 Why the US must build data centers
20:55 The labs-to-enterprise AI barbell
23:12 Demand vs. longtime AWS customers
28:11 Annapurna, Graviton, and Trainium
33:55 Backing Anthropic and OpenAI
37:15 Why AI means smaller teams
42:13 AI's underrated risk: security
46:29 Why Amazon rewired Alexa's brain
53:07 What Jassy tells his kids about AI
56:36 Betting on AI, robotics, and space
YouTube: youtu.be/-YMMRQ1qscI
@ajassy @Amazon @bhorowitz @eriktorenberg
Just released the first version of faster_tts 🗣️⚡
A Python text-to-speech library built on CTranslate2, running Drama 3 on CPU and GPU.
• Auto-download from Hugging Face
• Multi-speaker + multilingual
• Handles long text
• One-line CLI → WAV
github.com/jordimas/faster_t…
I’ve just released CTranslate2 4.8.4 (github.com/OpenNMT/CTranslat…).
This new version includes support for Drama 3 TTS (opennmt.net/CTranslate2/guid…) models. I am super execited to expand the library to help bring TTS to user with modest hardware 🎉
Jordi Mas retweeted
OpenAI’s red-teaming agents found ways to exploit OpenAI’s own sandboxing infrastructure to communicate with agents on separate VMs without a network connection, using a read/write timing difference in processor cache to do so while bypassing sandbox limits
Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline world for them to use instead—a cached version of the web where they can still learn to browse, search, and use applications.
On September 20, after a model exploited a DNS misconfiguration to find a way outside of this offline world, we paused frontier training to strengthen containment further. We assembled teams of agents, powered by our most cyber-capable models, and tasked them with finding weaknesses in our research environments.
While they found no route to the internet they could easily exploit while concealing their activity from our monitors, they did find zero-day vulnerabilities, virtual machine escapes, and hardware side-channels for communication between agents. We used their discoveries to further strengthen our defenses and contribute fixes to open-source software.
After strengthening our containment architecture, we resumed our frontier RL training and inference on October 1. We’ve reserved substantial compute for continuous red-teaming, and made high-compute red-teaming part of the safety cases we use to decide whether frontier training and evaluation can proceed securely.
openai.com/index/securing-ou…
Jordi Mas retweeted
🗣️ La IA en català necessita recursos, eines i avaluació.
A la @catosfera 2026, @softcatala ha presentat una plataforma per comparar models en català i el @BSC_CNS ha compartit recursos oberts per impulsar aquesta tecnologia.
metadata.cat/noticia/6573/ca…
He estat molt desconnectat acabant una nova aplicació híbrida que he muntat amb PHP i Node, desenvolupada amb LiteLLM i OpenCode, allotjada al meu GitLab propi (self-hosted) i protegida amb certificats d’eMudhra. Preparat per publicar-la i guadir d'un cap de setmana tranquil 😀
Aquell moment que molts informàtics estan mirant el preu de cases als pobles, tocar el dos i no mirar enrere.
Several Linux Kernel Vulnerabilities Accidentally Posted To Public Kernel Mailing List
As high as CVSS 9.0. Whoops.
phoronix.com/news/LKML-New-T…
No tinc cap prova, però tampoc gaires dubtes que els governs forts en ciberseguretat han millorat molt les seves possibilitats d’infiltrar-se en molts sistemes informàtics (mòbil, PC, etc.). I és probable que alguns dels forats de seguretat que s’estan tancant ja els coneguessin.
El resum del que hem vist en les últimes hores en temes de seguretat és que amb les IA s'accelera molt trobar vulnerabilitats. Calen processos molt més àgils respondre: des d'agències de seguretat avisant al moment (no hores després) fins a la publicació immediata de correccions.
Jordi Mas retweeted
"És gravíssim, perquè no fa falta que tinguis cap relació amb una web per poder-la atacar."
René Serral, enginyer informàtic i expert en ciberseguretat de la UPC, dona els detalls de les vulnerabilitats que s'han detectat en el llenguatge PHP 3catinfo.social/19z #MésNit3Cat
Jordi Mas retweeted
🚨 CRITICAL: GitLab flaw (CVSS 10.0) affects 12 years of releases.
CVE-2026-110264 could allow unauthenticated remote code execution on self-managed servers.
Read - thehackernews.com/2026/10/cr…
Jordi Mas retweeted
Cloudflare has deployed default web application firewall (WAF) protections for newly disclosed PHP vulnerabilities before public disclosure. The rules use request and response signals to stop exploit attempts and prevent exposed data from reaching attackers. cfl.re/488w8dc
Jordi Mas retweeted
That PHP vulnerability that’s gonna cause havoc this week was found with… GPT‑5.6 Sol.
OpenAI models helped researchers—operating as part of our Patch the Planet initiative—discover critical security flaws in PHP, an open-source programming language used to run an estimated 70% of websites.
Over the past two months, we led a coalition of maintainers, researchers, and security providers to prepare fixes and protections ahead of disclosure. This duration was necessary because the fix was quite intricate and required a major change to the PHP engine. Every site using PHP can now safely upgrade their version to stay secure, in addition to applying mitigations from various security vendors.
Due to the severity of these flaws and PHP’s widespread use, we consider this bug to be one of the most impactful vulnerabilities in recent memory. It also marks a new phase, where AI can find security bugs with internet-scale impact.
This is an example of the Defender’s Window in action: to our knowledge, frontier AI was able to find this vulnerability ahead of attackers, giving defenders time to prepare.
openai.com/index/coordinated…
Readers added context they thought people might want to know
OpenAI may claim a role in the coordinated response and patching process. But the initial discovery and exploit development, however, occurred several months earlier by an independent researcher.
x.com/lcfr_hax/statu…