Trevor Bingham
@22TrevorBingham
Followers
775
Following
373
Media
71
Statuses
4,251
AI is the most dangerous thing we have ever created. If we don't stop it will take our jobs and lead to massive AI arms races inevitably ending in war.
Joined March 2022
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
JIMIN JIMIN
• 1016585 Tweets
JIMIN IS COMING
• 429703 Tweets
ZEGNASS25 x GEMINI FOURTH
• 357415 Tweets
#Jimin_MUSE
• 214592 Tweets
PJM2 IS COMING
• 188748 Tweets
Mahalima Rocks AgusanDelSur
• 169464 Tweets
LA LETTRA IS COMING
• 168358 Tweets
JIMIN COMING SOON
• 146202 Tweets
Ucrania
• 127505 Tweets
Dudu
• 77926 Tweets
Romania
• 65508 Tweets
Nintendo Direct
• 65291 Tweets
The Truth Untold
• 62410 Tweets
音楽の日
• 55761 Tweets
ARMY ONLY
• 42083 Tweets
Leila
• 39672 Tweets
Belgium
• 30876 Tweets
Wayanad
• 29006 Tweets
Lunin
• 24079 Tweets
Unai
• 23609 Tweets
प्रियंका गांधी
• 23490 Tweets
ニンダイ
• 21620 Tweets
#HappyBirthdayHokuto
• 21603 Tweets
Brandy
• 21245 Tweets
Mudryk
• 20365 Tweets
#松村北斗誕生祭2024
• 18030 Tweets
Stanciu
• 17288 Tweets
Slovakia
• 16964 Tweets
$LEGION
• 13276 Tweets
Lukaku
• 12668 Tweets
Roumanie
• 12419 Tweets
KATY PERRY
• 11792 Tweets
Doku
• 11028 Tweets
Bélgica
• 10233 Tweets
Pinned Tweet
In a year or two, cutting edge AI programs will learn to do something truly amazing: They will be able to modify and improve their own programming as well as any human programmer. We will have entered the age of lighting fast AI self-improvement.
6
13
61
When do I have the right to destroy your way of life without your permission?
OpenAI, Anthropic, and some other companies are about to end our way of life. Right now we have no say in the matter, and very likely will never have a real say because we will wait too long to make
67
30
169
@JohnArnoldFndtn
@pmarca
@Atelfo
Three reasons for the confusion are too low a vitamin D dose, too short an intervention period, and occasionally, poor formulation or wrong type of vitamin D.
There are plenty of studies showing positive effects, though. Here is just one.
"Vitamin D3 Supplementation at 5000
6
7
89
We're at the OpenAI operating headquarters at 575 Florida Street, San Francisco. No company signs identify the building as their headquarters. They act is if they're ashamed of their name. They should be. They have betrayed their original nonprofit ideals. Plenty of time to come
12
12
84
@save_children
@paulg
Who reports those numbers originally? How are they verified? Finally, how much moral responsibility for this should Hamas be given for not allowing the timely evacuation of the combat zones?
32
6
72
The OpenAI protest today is set to go!
OpenAI doesn't have a right to push forward to AGI on their own. They're guilty of unprecedented intellectual property theft for their commercial benefit. AGI will be the most dangerous and powerful technology we have ever created. No
15
13
72
@ArmenAgha
When does the Meta board room drama start?🙂 Has Yann been fired yet?🙂 Who will play Emmett's part for the rest of the weekend?
2
2
67
AGI will be a strong net negative for people. AGI propagandists extol the benefits of AGI for the good things it will do for us, but either down play the negative things or never mention them. Hear are some some the negatives that are never honestly addressed.
1. AGI will
21
11
68
Huge respect for Sam Kirchner. Things are very serious, time is short and no one seems to care. Somehow we must get the people's attention. This needs to be reposted from one end of Twitter to the other. 👍
6
13
62
OpenAI and the other AI development companies do not have the right to open the door to AGI on their own. It is a far more important decision than developing the atomic bomb, and every person should have a chance to understand what that means in their own life and the life of
9
8
50
Still at Open AI operating headquarters at 575 Florida Street. 🙂 Still time to come out and the greet the good folk at OpenAI and tell them what you think about their work on AGI. We will be here till 6:00.
10
4
51
Pause AGI Development and Enjoy Current AI Technology. Sam Altman and others think there is a real possibility of “lights out” for the human race if we continue. Pausing to assess right now is smart strategy to avoid that possibility while still allowing us to enjoy substantial
11
7
50
2022 and 23 were history-making years in AI research. The world is about to change forever as a result. Sometimes great breakthroughs happen with much fanfare and attention, but very often they come in almost unnoticed at first. This is what happened in the last year. We lived
5
7
46
Made it! We're here in San Francisco for the World Pause Coalition protest at OpenAI headquarters tomorrow. 🙂 If you're in the San Francisco area come and join us. AI is going to affect everyone, and everyone has a right to make their voice heard on this. The more people who
7
9
46
The human bastion of music is falling to AI. Science is next. It may not be the most glamorous human domain of activity, but when it falls (soon) it will be by far the most dangerous advance of AI. Can't anyone see where this is going?
Suno V3 was already the most incredible tool for me, but this is another level.
2 voices singing in harmony, the sound mixed with great dynamics, all the instruments can be heard, impossible to distinguish from a recorded extract. Generated in a few minutes. It's breathtaking.
10
16
104
14
4
43
A lot of us see the danger of Super Intelligent AI, and we talk about it a lot here on X. It's a good way to let off some steam on the subject, but we need to do more than let of steam. If AGI is as potentially dangerous as we say we believe it is, we should be willing to do more
17
9
42
Couple hours ago leaving Tulsa.
Only about 24 hours to go. 🙂
Looking forward to seeing everyone who can come to OpenAI protest at their headquarters at 575 Florida Street, San Francisco, on Tuesday. Very important to come if you can. If you can't, please retweet.
4
12
37
@MadhubantiChat3
@libertycopula
Why were the "brave" Hamas fighters hiding in a hospital? You need to condemn them for this. Everyone needs to call them on their cowardly behavior of hiding behind women and children.
5
6
37
Have you thought about what the world security situation will look like right on the edge of AGI? According to the only real experts in this very new field (that is, those who are actively building it or have recent experience working on it) there is a fairly good chance AGI is
11
7
39
@sama
Sam Altman: "short timelines and slow takeoff will be a pretty good call i think, but the way people define the start of the takeoff may make it seem otherwise"
So in other words, from the perspective of most people, short timelines and a short takeoff. That is your safest bet,
6
2
34
@Simeon_Cps
I am still guessing that their current state-of-the-art in-house AI version, whatever it is called, is surprising them with emergent ability beyond their expectations.
1
1
32
How close can a lab get to AGI before they are involuntarily paused?
Assume that company X is has a broad based, say 100-Billion-dollar push to full AGI. What happens when they get really close to turning on the full AGI model with all that massive compute behind it? Assume that
16
8
30
@romcabrera
@fchollet
@liorb_d
Exactly. He knows that too. It was a classic Francois Chollet red herring argument: only AI will save us. 🙂
1
0
30
@thealexker
@Scobleizer
Relatively small but good example of how advanced AI is going to devastate our economy and put a lot of software engineers and everyone else out of work. Wait for GPT-5. It will be a lot better/worse. Wait. 🙃
4
2
29
Very interesting that Eric Schmidt is starting to worry a little. He needs to be much more worried and he needs to have the courage to show it in public. Bad times coming if we keep going.
Eric Schmidt: the point at which AI agents can talk to each other in a language we can't understand, we should unplug the computers
158
94
488
4
4
27
It will cost us a lot, but we can prevent AGI. Current multimodel AI is capable of watching (and listening to) everyone, everywhere, all the time. Sounds very dystopian and it is. But we all (well, OK, Beff, almost all of us🙂) will shortly realize that to survive we have to
9
4
27
Why should we be afraid of AGI?
AGI leads very quickly to ASI because it will have world class expert human level AI development capability built in to it. The critical difference between human AI developers and AGI driven AI development is speed. The AGI model will be able to
15
6
25
There's an additional major danger that most people are overlooking. If one country is about to put the whole human race at risk of a 50% chance of extinction, how do you think the other nations of the world will react?
Meet the guy who invented
@OpenAI
's RLHF technique ().
His name is Dr.
@PaulFChristiano
, PhD Comp Sci from UC Berkeley.
Paul says there's a 50% chance of doom once AI systems are human-level intelligent.
So do we keep training smarter AI systems or no?🤔
73
157
789
8
2
23
@mezaoptimizer
I think on the corporate level there are a lot of people who are thinking about a post AGI world.
Sam Altman understood the situation correctly at least 4 years ago when he mentioned the possibility that OpenAI might “maybe capture the light cone of all future value in the
2
1
24
A perfect alignment solution for AGI does not solve the problem of looming AGI supremacy. In fact, guaranteed safe alignment makes it more sensible for a nation to try for AGI supremacy if they know they can control the AGI in the long run. This increases the chances of an AGI
10
4
22
Back from the PauseAI protest on Monday. Congratulations to Holly Elmore (
@ilex_ulmus
) and
@pauseaius
for organizing such a successful protest.🙂👍 They had a surprising amount of media coverage. This is what counts. The average person needs to see it on TV to believe there is a
2
2
22
Simeon is coming to a similar possible conclusion to Joscha Bach's but from the other side of the AI safety spectrum. I hope they are both wrong, but there is a real possibility that AGI is closer than our most optimistic estimates. We need a little more time, even 6 months
We're now in the worrying timeline where we could wake up tomorrow and discover that the fast takeoff has started
28
20
146
3
0
21
Here is a fascinating look inside the thinking of one of Beff Jezos' faithful e/acc acolytes:
"The downside to AGI is finite. The upside is infinite."
E/acc people convict themselves with everything they say.🙂 The down side in their mind is the extinction of the finite
3
1
21
Blocked By LeCun.🙂
While I don't know for certain, I suppose it was for "insulting" language. 🙂 There is some truth in that. Describing some great public mental and moral lapse in someone is bound to be seen by them as insulting and while it was not my purpose to insult the
4
1
21
@ErrataRob
@JohnArnoldFndtn
@pmarca
@Atelfo
I am not basing my comment on one study.🙂 The sample study I provided was just to show there were studies showing beneficial effects. Because of the many possible variables, even experienced researchers often have contradictory opinions on these things in the early days. With
0
0
20
Musicians are well on the way to being "solved," even the most talented. First they came for the artists and actors. Then the they came for the musicians. They are literally going to take your job away next.
When are we going to wake up, people?
It is time to stand up and be
Music well on its way to being solved by AI
It's nice if this eases the production of good music, but is this something we really needed to automate?
I suspect there are far more talented musicians than can earn a living doing it now — it's a pity that this problem will worsen.
14
3
21
10
2
19
@secretsqrl123
He is not really a bad a guy. 🙂 It was political. He was being loyal. I could see that from the way he talked about the "SecDef." They made a sudden U-turn and he got left behind. The main thing though is to get the help to the Ukrainians they need right now.
1
0
20
No corporate identification, but we're at OpenAI operating headquarters. Plenty of time to join us.🙂 Interesting thing, most people walking by don't know it's OpenAI headquarters. Even neighboring businesses didn't know. Their plan was to hide their presence. Don't let them get
2
1
19
"I think it's pretty likely the entire surface of the earth will be covered with solar panels and data centers..." Ilya Sutskever
Anyone who works actively or passively for such an outcome is an enemy of mankind and of the earth. It is almost time to draw lines in the sand. What
7
0
20
@klarnaseb
"We’re spending less on photographers, image banks, and marketing agencies." This will be more and more true as we get more powerful AI. It will eventually include the managers and owners too, wait for it. How the owners? With advanced AI managers, owners will not be able to
0
1
20
This article is good. Really good. It is worth a read. Torres has Beff Jezos and the rest of e/acc crew's number down cold. Can't recommend this article enough.
7
5
19
Anthropic is wrong to allow Claude to hallucinate about itself and it should be liable for what it has done. The garbage Claude is spouting about itself is actually messing with people’s minds. "Wow! Maybe it really is alive and conscious and suffering as it claims
12
6
19
This graphic gives direct insight into these people's credibility. It exposes some as truthful and some as liars. That is because AI will without question cause massive job loss. That is a given that everyone in the field understands. Some people publicly pretend it won't. They
Will AI ultimately replace all jobs? Lots of smart people disagree – what do *you* think?
@sama
@ylecun
@elonmusk
@MelMitchell1
90
20
184
6
1
18
The leaders of the companies developing AI see the job losses coming just as clearly as we do, but it is in their own interests to pretend they don't see any thing.
@AISafetyMemes
That is one of the things that made me instantly alarmed when ChatGPT got out, it obviously will lead to chaos due to job losses, it's so straightforwardly obvious, that I can't believe people in the AI community who are known for their high intelligence just can't see this.
3
0
18
1
1
17
@Zvbear
@wil_da_beast630
Keep in mind that the current version of covid is significantly milder and that the average person has some immunity of one sort or another to covid in general. Also the some of the most susceptible peoples have already caught it and died.
3
0
17
@fchollet
You gloss over the remarkable ability of what AI can do. Yes, they can't extract information that is not there, but they can process larges amounts of information for very small cues that would take humans a very long time and then find useful patterns in that information faster
1
0
18
What I got wrong about full AGI:
I was wrong to say full AGI would be as smart as the best humans. It will be smarter. I was wrong to say full AGI would be a thousand times faster. It will be over ten thousand times faster. What this means is that AGI will quickly move from
7
1
17
E/acc people are determined to push their version of the future on us whether we want it or not. They don't care about what other people think and they don't care what their desired future would mean for the developing world.
@22TrevorBingham
Powerful won't care about weak people opinion. AGI is just around the corner, you will witness it. Not only AGI but also ASI. You will be post Singularity being.
2
0
3
3
2
17
@ESYudkowsky
Why anthropomorphize the situation? It is nothing more than a very complex program running on a powerful computer cluster. However, there is still a good lesson here. The real problem will be us humans using and abusing AI for our own ends, not so much Skynet going rogue. No need
3
0
15
High level AI is drastically changing the time lines for human level robots. There were two problems to be solved before we could develop relatively low cost but highly capable robots. The easier problem was the hardware problem of building a robot that could walk, pick up things
The timeline split of AI vs Robot Hardware has changed
the last 90 days i’ve witnessed industry leading AI in our lab running on humanoid hardware, and frankly it’s blown me away
i’m watching robots performing complex tasks entirely with neural nets. AI trained tasks that i
149
381
2K
1
2
16
"How is OpenAI not shut down yet?" Partly our fault. We need to try to jumpstart a much louder public discussion on the dangers of AI and the need to Pause now and Asses where we are going. Maybe several protests in succession could build momentum on this issue. I know that not
@daganshani1
How is OpenAI not shut down yet?
If not for the fact that they are playing Russian roulette with the entire Earth, then for the fact that they stole Internet-worth of value from all of humanity..
5
1
20
2
0
14
@teortaxesTex
@mezaoptimizer
The prize is indeed great. But it is a lot bigger than "just" a new Renaissance. The real prize is SI supremacy. Maybe it is not your cup of tea, but it will be the great prize many people in top positions in government and business compete for.
Any AGI/SI system that is so far
5
2
14
@new_community
@FiorellaIsabelM
He goes up a notch in my mind for not being afraid take a stand that is unpopular. To me, this means that his vaccination stands are not one off, but that he really has the courage of his convictions. How old fashioned! How refreshing.👍🙂
1
2
16
There is no future without a Pause. Without a pause, AI competition will turn into an AI arms race which will lead to preemptive wars. If your opponent is about to develop AGI+ technology, you won't have any choice but to preemptively strike them before they reach ASI supremacy.
0
1
15
@DwayneCodes
@geoffreyirving
Those 650 people are not working close enough to him to know the real Sam. Several years ago the OpenAI workforce was much smaller. Even then, from my experience in a different situation, only a few dozen at most will really get to know the leader on a personal basis.
0
0
16
@fasc1nate
This is so interesting.🙂 If they can remember that well and react that fast, it seems to me that they can be trained to manifest much higher intelligence than normal.
2
0
12
February 29 to March 13???? TWO WEEKS To "embody AI"??? If this isn't fast, hate to see what a "fast take off" looks like.
"Excited to partner with OpenAI and Microsoft to bring embodied AI to the real world"
In addition, OpenAI & Figure signed a collaboration agreement to develop next generation AI models for humanoid robots
The collaboration aims to help accelerate Figure’s commercial timeline by enhancing the capabilities of humanoid robots to process and reason from language
5
13
230
4
4
16
AI will soon be able to do engineering. This will change everything, and not only for the good. As little as a year ago this was not obvious to most people.
"So, is it possible for engineers to be replaced by AI-driven systems? Yes, but it's not likely to happen anytime soon.
1
3
16
Here is an excellent thought from the Executive Summary of the Helen Toner et al paper that irritated Sam Altman and exacerbated his rift with the board.
"AI technologies are evolving rapidly and enable a wide range of civilian and military applications. Private sector
4
1
15
Join us Tuesday morning at 10:00 in protesting Anthropic’s reckless plan to quickly expand the power of its best AI models to compete with OpenAI in the commercial AI market.
Anthropic came into being supposedly out of concern over the commitment of OpenAI to safety. Now it is
0
3
15
This is really Good.
"1. Probably there will be AGI soon -- literally any year now.
2. Probably whoever controls AGI will be able to use it to get to ASI shortly thereafter -- maybe in another year, give or take a year.
3. Probably whoever controls ASI will have access to a
Daniel Kokotajlo (OpenAI Futures/Governance team) on AGI and the future
10
28
134
2
0
15
@adcock_brett
@ShepJoel
@Figure_robot
If you do the work of actually thinking this out, no one is going to want a human worker if they can buy an equivalent robot for $150,000 or less. Massive social disruption ahead of us from only this single effect of AI and robotics.
2
1
14
This is really an excellent analogy of the situation we face right now. Too many people are afraid to take the implications of recursive improvement seriously. There are too few Einsteins who can quickly grasp an unfamiliar idea and are not afraid to act on their intuition.
There’s a striking similarity between physicists hearing about nuclear chain reactions and AI researchers hearing about recursive self-improvement.
Key bottlenecks in both cases include willingness to take high-level arguments seriously, act under uncertainty, or sound foolish.
7
48
415
1
0
14
Next generation multimodal AI will be able to do amazing things for us, but will not in itself be an extinction threat. If we pause now, we get to keep all the benefits we have coming from current AI. If we throw the dice again and move to AGI, we risk losing everything.
4
0
15
"Workers Fear For Their Jobs." They have only the faintest glimmering right now of what is coming. But you AI companies and executives have no idea of the overwhelming revulsion and anger that is coming your way in a year or two.
4
4
11
@World_At_War_6
@sentdefender
The truth does matter in the longer run. They need to get the information out. Were there any Israeli airplanes over Gaza at that time? What targets did they hit, if any? They need to get that information out. If no Israeli jets were over Gaza at the moment, they need to say that
3
0
14
@kosh516
@sama
Sam's only chance of achieving ASI is to do it unexpectedly soon. To be known to be close to ASI is not a safe position to be in. He knows this. You need to move very fast to survive. If he does succeed (I don't believe he will), he will become a "god" in terms of his own power
3
0
15
@AaranTheArtist
I am very sorry. But everyone should realize that what will happen to your job shortly will be happening to most of us next.
3
0
13
AI that can write software. Not new, but one to best so far. Faster and faster we go. We can all see where this ends, right?
Today we're excited to introduce Devin, the first AI software engineer.
Devin is the new state-of-the-art on the SWE-Bench coding benchmark, has successfully passed practical engineering interviews from leading AI companies, and has even completed real jobs on Upwork.
Devin is
5K
11K
45K
3
1
13
E/acc advocates want to gleefully plunge into every new AI technology with no restraints or regulation. Their childish enthusiasm for powerful unregulated AI will cost all of us a lot, beginning with our freedom and quite possible ending with our lives.
You wanted full
3
3
14
AGI is not likely to happen, but for reasons that most people have not yet considered carefully.
First, though, there should be no doubt in anyone's mind that we could develop AGI. Because of their relatively narrow tech focus, AI developers are some of the worse possible
9
2
12
It was always going to go this way. They will take all the e/acc dreams and twist them into a never ending nightmare.
1
0
14
The labor movement in the United States will be dead in 3 or 4 years regardless of whether we go to AGI or not. Smart, capable robots will be here later this year. There is going to be a massive replacement of human workers with intelligent robots. Labor unions have played a
3
1
14
Mining and manufacturing are going to take off to a degree we can't imagine right now. The big AI and computer companies are already demonstrating they have a tremendous local political clout. They are going to rape and abuse the environment even more than they have IP right
The conversations I'm having with hardware and big tech firms, and Fortune 500 data/IT leadership are getting much more serious about prepping for AI.
Ppl have no idea how much of the world's GDP is going to be in:
-Mining metals
-Power /energy
-AI hardware
-Datacenters
-Data
18
13
131
2
3
13
Some people see where we are going:
"The world system is steaming towards AGI at the limit. Everyone who has accepted the realization that this will be the last invention and has the resources to take part will do everything they can to help shape the final moments." Translated
2
3
12
The big danger with near AGI LLMs is that they will almost certainly be able to quickly improve their own code. This means they will be able to quickly move from near AGI to AGI to AGI+. Things are lot more dangerous than most experts realize. After a life time of hastening this
1
1
13
We will soon face a new national security threat from powerful AI that can be used to quickly increase knowledge in basic science and technology.
Specially trained hybrid AI models will soon be able to push forward the frontiers in science and engineering with minimal human
5
5
13
@geoffreyhinton
There's been a lot of progress made in training techniques. 100x increased computational training resources may add up to more than a hundred x improvement. We may be less than 18 months from AGI. I think this is both a real technical possibility and a beneficial strategic
3
1
12
This is really important:
"Got it. How long do we have? This is uncertain. It might be 6 months or 5 years. Idk. Open Source is a bit behind on compute but not that behind on techniques. If the bottleneck is data or technique rather than compute, we are fucked." Technique is
We might be dropping the ball on Autonomous Replication and Adaptation.
Here is a little Q&A
Can you explain your position quickly?
I think autonomous replication and adaptation in the wild is under-discussed as an AI threat model. And this makes me sad because this is one of
12
10
91
0
1
12
@MikePFrank
@sama
"it’ll be a slow takeoff."
There reasons to think it will be sooner than we expect. OpenAI has said ASI this decade. Sam Altman said this not to long ago:
"building agi is a scientific problem building superintelligence is an engineering problem"
We
building agi is a scientific problem
building superintelligence is an engineering problem
463
510
5K
3
0
12
Advanced AI research is the single most important issue facing us. The government has the obligation to act to protect the people. I think there is a chance they will act in time, but there is no proof right now that this is so. Until we see decisive movement on the part of the
They need to be forced. By law. The police or even army need to go in if they don't comply. This is what would be happening if the national security (aka global extinction) threat was taken seriously.
23
5
62
2
0
12
@petemandik
@ESYudkowsky
The disturbingly poor analogy. Even good analogies are useless for proving an argument: they can at best only help explain it.
Too many bad analogies being used as proof for various positions.
0
0
10
“What is "AGI"?” In the context we are most concerned with, AGI is AI that can equal the very best human scientists and engineers. It is AI that can play the game as good and even a little better than the best human players.
We take it for granted that our current sub-AGI can
1
2
12
@BasedBeffJezos
They were always going to regulate the common use of AI. The deeper question is, who is going to regulate government use of AI?
1
1
11
@JonLamArt
YES. I believe we are going to see a number of new FTX type scandals in the next few years from AI companies. Some people today who are almost worshiped as "Founders" will be keeping company with Sam Bankman-Fried in the future.
1
0
11
We don't need AGI to have a wonderful future.
“We have been able to just accelerate the discovery of antibiotics,” de la Fuente said. “So instead of having to wait five, six years to come up with one candidate, now, on the computer, we can, in just a few hours, come up with
1
2
12
@dr_park_phd
"I want our mindchildren [AI] to live even more."
We have to realize that this is a deep seated problem with some people. Regardless of anything they say, they are first of all traitors to the human race. You may not like that, but it is true, and the sooner all normal people
0
1
11
A few very wealthy people have decided to change everything we all take for granted in the world for better or very possibly for worse. Human level artificial general intelligence (AGI) is coming in a few years because a small number powerful corporations like Microsoft and
2
3
11
@OpenAI
Mostly public relations. "We are pressing on with the creation of AI systems that may destroy humanity but we want everyone to know that we are starting a preparedness team to help protect everyone from the possible catastrophic dangers of our AI work."
3
0
11
@ylecun
@MelMitchell1
@tegmark
AI research and development does pose a very large threat to humanity. Extinction is an artificially high bar. There are plenty of lesser but still catastrophic outcomes.
One such catastrophic outcome develops from a particular country being a ahead of everyone else on the
2
2
10
Near AGI engineering and science models will soon be dangerously capable. We need to stop and think about where we're going. Things are moving fast. Specially trained AI development models and AI general engineering models will be dangerous before AGI. Specially trained chip
5
3
11
@matthew_pines
It shows what everyone already knew: The AARO report was a total fraud. Which is clear proof that that the subject is considered important enough to justify public lies. Unless our survival is at stake, nothing justifies public lies of this magnitude.
0
0
11
We can make a strong pause work if we want to.
First, humans will do what they have to do to survive when the stakes are high enough.We will do what we have to do to survive, even things we don't want to do. If we decide that is the only way to survive, we will hold our nose
1
0
11