¿Entendéis que nos podríamos plantar con Astra y parar el desarollo de la IA como humanidad y que aún así tendríamos años de beneficios y avances radicales con lo que una IA así podría aportar en su integración en la sociedad?
Y que sin embargo es sólo un peldaño más de una escalera que va a seguir creciendo y creciendo aún más en los próximos meses.
De aquellas inversiones locas en computación de la que hace un par de años mucha gente se burló todavía queda mucha capacidad por ser habilitada para que los grandes labs puedan seguir investigando (y soportando la creciente demanda). Las técnicas de Deep Learning se seguirán mejorando. Los datasets seguirán creciendo y puliéndose. Los simuladores y entornos de entrenamiento seguirán refinando el entrenamiento de los modelos. La sociedad continuará integrando más y más la IA en todos los procesos. La ciencia se acelarará con la revolución en matemáticas y otras areas que estamos empezando a vislumbrar. etc. etc. etc.
Disfrutad del paseo, será un viaje fascinante.
there are a lot of benchmarks that suggest 5.6 sol is the best model in the world right now, but the most reliable way to tell is that elon is obsessed with me again
Georg Ohm's foundational principle connects voltage across a conductor to the current it carries through its resistance.
Starting from the electric field integral defining potential difference and current density over a cross-section, the relations simplify for constant fields and uniform materials to V = RI.
Straight-line graphs illustrate constant resistance in ohmic components such as resistors, while curved plots reveal changing resistance in non-ohmic elements like diodes.
Circuit designers apply the law to compute required resistances and currents when building safe and efficient electronic devices and power systems.
I made a personal black hole that makes you take breaks 🕳️
A shader for Ghostty that spawns a small black hole in your terminal - it drifts around, gravitationally lensing your text. The longer you work without stopping, the bigger it gets, until it's basically demanding you go touch grass
Take a break and it quietly shrinks away
Stanford proved that ChatGPT, Claude, and Gemini are all secretly running at a fraction of their real creative capacity.
And one prompt unlocks the version they hide from you.
This paper reveals that the multi-billion dollar process of "Alignment" (RLHF) has accidentally lobotomized AI creativity.
Researchers discovered a phenomenon called Typicality Bias.
When humans rate AI responses, they have a deep psychological drive to choose the most "typical" or familiar-sounding answer.
They don't want the most creative story; they want the one that sounds most like a generic story.
The AI learned this.
It realized that being truly creative actually hurt its safety and preference scores.
So it entered a state of "Mode Collapse", it effectively hid its most original ideas to stay within the safe, boring boundaries we set for it.
But the creativity is still there. It’s just locked.
Stanford researchers found a "master key" to bypass this training and it is ridiculously simple.
They call it Verbalized Sampling (VS).
Instead of asking the AI for one answer, you ask it to verbalize a distribution of responses and their probabilities.
Ex: "Generate 5 unique jokes about coffee and the probability that each one is actually funny."
The results are staggering:
- 2.1x increase in output diversity.
- 25% jump in human evaluation scores for creative writing.
- Zero loss in factual accuracy or safety.
By forcing the model to calculate its own probability distribution, you "unlock" the 66.8% of generative diversity that was suppressed during training.
@sama I built a fully integrated cybersecurity cross correlation engine with a ChatGPT over lay to "talk to" data. It allows many cybersecurity tools which normally cannot cooperate to perform data fusion. It's amazing and 5.5 made it possible. #cybersecurity#openai
@LuisinaLuca Pues Cleopatra estaba tan enamorada que se fue a la guerra por amor a Marco Antonio sabiendo que tienen todas las de perder. Y luego de perder se suicida jajajajaja