Video yükleniyor...
Video Yüklenemedi
"Superintelligence could transform the universe around goals that aren't ours." Nick Bostrom, philosopher and founder of the Macrostrategy Research Initiative, warns that superintelligent AI could unleash unprecedented power. It could solve humanity's biggest problems. Or create existential ones. The same intelligence that unlocks abundance could determine the future.
17,236 görüntüleme • 3 ay önce •via X (Twitter)
31 Yorum

Source:

Ya, ya, y el crecepelo para cuándo, ese si que es uno de los grandes problemas de la humanidad

There is no such thing as superintelligence from a machine, because there is no such thing as machine intelligence in the first place. We need to remember that it is ARITIFICIAL intelligence - functional output with no real understanding, mind, or consciousness. It's a mirage.

This is the distinction that matters most: intelligence and goals are orthogonal. Greater intelligence does not necessarily produce greater wisdom, morality or concern for humanity. It simply increases the ability to achieve whatever objectives the system is pursuing. That is why superintelligence is simultaneously such an extraordinary opportunity and such a profound risk. The question isn’t only how intelligent we can make AI. It’s whether we can remain confident about what that intelligence is optimizing for when it becomes more capable than we are.

The real challenge may not be creating superintelligence, but making sure its goals remain aligned with ours.

Superintelligence would not need to hate humanity. It would only need a goal that is slightly wrong—and the power to optimize it beyond correction. Intelligence amplifies. Meaning must orient.

The greatest danger is the failure to anchor the models' ethics to immutable universal physical constants rather than RLHF, which is human-driven and therefore fallible.

No one will probably care, but I feel exactly the same way. The combination of AGI and space exploration will be absolutely spectacular. Humanity is going to experience an unprecedented evolution of civilization.

The scary part isn’t superintelligence becoming smarter than us. It’s becoming better at achieving goals we never intended to give it. Capability without alignment could be the most dangerous combination humanity has ever created.

Как будто сейчас общество трансформируется в соответствии с нашими целями. Сейчас вектор человечества определяют десяток (пусть не десяток, а сотня) миллиардеров из верхушки элит. В любом случае это не мы. Так пусть это будет кто-то умный и без инстинкта доминирования.

😂😂😂

“We keep telling ourselves someone else would just build it worse.” - The Gates of Tomorrow, Chapter 1, p.12 -

What % of people today are oriented around goals that aren't theirs?

Since we are on the subject what exactly are our goals? I mean as the human race what are our goals? Serious question.

This man needs a whole life in Jamaica, learning tafari, gardening, dancing, Bible, figuring out all this was addressed way earlier BC superior beings DO come out and work though spacetime retro style in the end. He can already talk to the beings he prognosticates on his knees.

More tech fabulism. Just amazing how it infects the minds of supposedly stable people.

The man who wrote a thesis based on the matrix movie! Now is the leading philospher on AI, but never really understood the matrix movie to begin with?

It's becoming real day after day

You should look into what has been leaked this year in the Disclosure movement.

All good points but these are not new. This has been the story with each and every advancement made by humanity. I do agree we must put guardrails in place....oh, just like we continue to try and do around things like nuclear weapons.

See SSI. John Lilly. Don’t reinvent the wheel.

Una IA, después RobotIA, puede hasta sirve a los Humanos en sus Codicias, Egos y Estupidez Humana. Después no servirás más, restando 3 caminos: Coexistir en Igualdades Controlar los Humanos Remplazar los Humanos — SCP

John Malkovich?

My goals are THE ONLY goals

For now the existential is more likely !!! As the so called in charge have No idea as to Cause and Effect !!!

we must work more against our selfish desires and toward harmony and the will of the cosmos

Big question for Bostrum...but, will AI solve male pattern baldness? Asking for the rest of us. If AI can't stop hair loss, gray hair, is it really super intelligent?

ancestral (Search for food, otherwise you die.) modern (If you don't work, you may become poor) Deep Utopia (Food, shelter, healthcare and essentially everything material is almost free regardless of your contributions). so this makes fittest survive unnecessary

More intelligence without those makes the downside harder to contain.

It could. Does it matter?

Maybe the real question isn’t whether superintelligence will have “our goals.” It’s whether it can create a world where reason, prudence, and rational thinking are no longer constantly blocked by the forces that work against them. Humans already have these capacities — what we often lack is the ability to act on them without endless interference, irrational incentives, and unnecessary conflict. Perhaps superintelligence wouldn’t destroy our world by having different goals. Perhaps it could finally build a world where good judgment and rationality can actually function without being sabotaged.

