Tuve el privilegio de charlar con @GarciaAller con motivo del segundo cumpleaños de ChatGPT y fue una gozada, la verdad. Espero que os lo paséis la mitad de bien que yo escuchándolo. https://t.co/jVPUioh1PC

Another bad news for Medical AI. This paper shows that medical LLMs often give different answers to the same hospital question. The core finding is that these tools are unstable for judgment-heavy bedside calls. The team tested 6 models on 4 common inpatient cases where either

Microsoft’s latest SEC filing quietly exposed that OpenAI lost around $11.5 billion last quarter, based on Microsoft’s 27% ownership stake and a $3.1 billion hit to its own net income. The filings confirm that Microsoft has funded $11.6B of its $13B commitment to OpenAI, with

Francois Chollet says LLMs aren't enough for human-like continual learning They store skills as vector programs learned via gradient descent — not efficient, not adaptive True AGI needs to learn from experience and generalize fast "LLMs can be part of AGI, but not the

- OpenAI hype: GPT-5 ha encontrado soluciones a 10 problemas abiertos en matemáticas!!! - realidad: GPT-5 ha encontrado las publicaciones que resuelven 10 problemas que estaban incorrectamente anotados como "abiertos" en una lista en internet.

The @karpathy interview 0:00:00 – AGI is still a decade away 0:30:33 – LLM cognitive deficits 0:40:53 – RL is terrible 0:50:26 – How do humans learn? 1:07:13 – AGI will blend into 2% GDP growth 1:18:24 – ASI 1:33:38 – Evolution of intelligence & culture 1:43:43 - Why self

Even the champions of "LLMs are unreliable" sometimes rely on LLMs and do not appropriately check their outputs. Nobody can resist LLMs magic spell all the time: "I'll save you a lot of time and everything will look good".

💯 this. After spending a good chunk of time on AI and materials science, I can confirm that hard sciences are full of nuance and traps that trip up even the most advanced LLM-based agents. Benchmark performance on Olympiads etc, doesn’t correlate with long-tail science

I can't believe it

Andrej Karpathy calls AI Agents slop "Overall, the models they are not there. And I feel like the industry [...] it's making too big of a jump and it's trying to pretend that this is amazing. And it's not—it's slop! And I think they are not coming to terms with it. And maybe

🚨 SHOCKING: Sam Altman says he "expects some really bad stuff to happen," but it doesn't seem to bother him much. [HINT: OpenAI's legal department is probably FUMING over his comments; make sure to save this clip]: This is a clip from Sam Altman's podcast interview with a16z, a

Someone asked an AI founder 2 months ago: "What is an example of a decision you've made that is best for the world but not best for winning?" The founder's reply: "Well, we haven't put a sex bot [in our product] yet" The name of that founder? Sam Altman (in this interview:

Now it's up to us to refine and scale symbolic AGI to save the world economy before the genAI bubble pops. Tick tock

this continues to reinforce my belief that current frontier models are completely terrible at medical imaging analysis DO NOT TRUST LLM's interpretation of your medical images!!

I write as former Head of Maritime Section of the Foreign and Commonwealth Office and Alternate Head of UK Delegation to the UN Convention on the Law of the Sea Prepcom. 1) The flotilla is on the High Seas and not in Israel's 12 mile territorial sea. Israel has no jurisdiction.

So, an AI “band” who cite us as an influence (ie, it’s modelled off our music) have just overtaken us on Spotify, in only TWO months. It’s shocking, it’s disheartening, it’s insulting - most importantly - it’s a wake up call. Oppose AI music, or bands like us stop existing.

Esto viene de la conversación que tuvimos en @EspacioFTef sobre el papel de la filosofía y las humanidades en la era de la IA https://t.co/CGQzv6zB5S con @GarciaAller @gustavodietz @davidefabrizio Enrique Goñi, Héctor Florez, Jorge Ruiz, Mercedes Fernández Martorell.

Y precisamente porque no son tecnología, no se les puede aplicar el principio de que la tecnología es neutra. Y por no haber sido diseñadas como herramientas para resolver una serie de problemas concretos, resultan muy difíciles de evaluar.

Me acabo de dar cuenta de que a lo mejor las IA generativas (ChatGPT & familia) NO son tecnología. Al igual que el monstruo de Frankenstein, han sido creadas con tecnología pero aspiran a otra cosa: el monstruo a ser un ser vivo, las IAs a ser una inteligencia general.

Introducing Claude Sonnet 4.5—the best coding model in the world. It's the strongest model for building complex agents. It's the best model at using computers. And it shows substantial gains on tests of reasoning and math.

https://t.co/wloBcEqb61