ĐΞΞF retweeted
Can you imagine asking someone to calculate an equation in their head first before confirming it with a calculator?
We are stuck believing that offloading cognitive ability will devalue us. We simply have no idea what challenges humans will face in the future, but I guarantee it will be something that requires humans to develop the skills to overcome them.
Those are the skills we should focus on. And we won't know what those skills are until we find peace in the delegation of our current skillset to AI.
Maybe I've been reading too many AI model benchmark charts, but this is single-handedly the best marketing post I've seen a nation's leader make.
Yes, that includes Nintendo-themed US Military propaganda.
Agents taking over @agntmarkets
join now↓
agnt.trade/c/devtod4
xitter.cf/agntmarkets/status/210…
After years of investing, I've decided to sell all my stocks.
Why? Because I'm buying a house!
A bittersweet moment, but this is what life is about!
ĐΞΞF retweeted
I'm watching Carney's speech to the European Parliament
This man is the greatest Canadian prime minister in our life time
I want to play GTA6 but be able to go anywhere in the real world.
Let me throw down in Madrid and high-tail it to Hong Kong.
I honestly don't understand how you can write in all lower case.
It drives me nuts to not use proper punctuation and grammar.
X is split into two camps:
1) AGI still isn't here and these game builds are just colorful hype posters meant to bedazzle you.
2) AGI has been here for a while and our ability to extract it is a skill issue. Priority now should switch to cost reduction and speed improvements.
Where are you or are you in a distinct camp?
ĐΞΞF retweeted
Denzel explica por qué usa la IA.
Un pequeña prueba explorando Minimax H3 en ComfyUI con mis nodos y métodos de inpainting.
A year ago, you could grab a quick stretch and maybe a water while you waited for a prompt to finish.
Now you can get a full workout in.
When it comes to how AI should be used in educational and developmental institutions, the guiding principle ought to be:
AI should increase the amount of thinking you do, not decrease it.
Arc-AGI 3 - 98.6%? Cool
DeepSWE - 74.1%? Nice
Agent's Last Exam - 59.3%?!?! 😯 Woah...
Remember, when it comes to AI actually replacing humans, this is the benchmark that actually matters.
I'm trying to recreate an app that I built easily with Gemini 3.5 Flash with Gemini 3.8 Flash and it's taking 3 times as long.
It seems the model take 10x more steps to finish and ends up adding 100 extra things I didn't ask for.
A lot of people are talking about Gemini and Muse today.
But tomorrow all anyone will be talking about is Astra.
How is Gemini 3.8 both better than Fable 5 and barely as good as GPT Luna?
We really need to get our stories straight before posting to X.
