Nobel Laureate’s AI warning

Trickster · Oct 15, 2024

“Last week, the Nobel Prize in Physics was awarded to two scientists — John J. Hopfield and Geoffrey E. Hinton — for research on computers that led to current AI technology like ChatGPT.

Hinton is often referred to as the “godfather” of artificial intelligence.

He is also outspoken about the serious risks AI poses to humanity — even leaving a high-profile role at Google so he could openly discuss his concerns.

Here’s some of what Hinton has said about the dangers of artificial intelligence:

Artificial intelligence technology could “amplify social injustice, erode social stability, and weaken our shared understanding of reality that is foundational to society.”

AI makes “it much easier for authoritarian governments to manipulate the electorate with fake news.” We need “legislation making it illegal to produce or share fake images or videos unless they are marked as fake.” (We already do this with currency.) But in the U.S., “one of the major political parties has tied its fate to the successful propagation of fake news.”

Combat robots powered by artificial intelligence could “allow rich countries to invade poor ones more easily.” And military systems that use AI to automatically set and kill targets could make wars impossible to control.

There is no one more qualified to testify about the ways artificial intelligence could be misused or get out of control. And of course Hinton is hardly the only expert warning about the threat of artificial intelligence — many of the world’s leading AI technologists are expressing concern.

As a society, we need to put strong guardrails on artificial intelligence and on the for-profit companies that are already locked in an AI arms race. But it’s no secret that elected officials are often too slow in keeping up with evolving technology — and that they are very much influenced by Big Tech.

That’s why it is essential for the American people to demand that Congress do more to grasp the risks of artificial intelligence — and to address those risks with comprehensive legislation and regulations that create guardrails to protect us — before it’s too late.

Tell Congress:

No less an expert than Nobel laureate Geoffrey Hinton — the “godfather of AI” — has been sounding the alarm about the profound risks artificial intelligence poses. Those risks are so serious, and the systems are evolving so rapidly — in part because major corporations are already locked in an AI arms race — that there isn’t time for the government to do its usual bumbling with new technology. The American people need you to fully grasp the threat of AI and to pass comprehensive legislation to protect us.

Last week, the Nobel Prize in Physics was awarded to two scientists — John J. Hopfield and Geoffrey E. Hinton — for research on computers that led to current AI technology like ChatGPT.

Hinton is often referred to as the “godfather” of artificial intelligence.

He is also outspoken about the serious risks AI poses to humanity — even leaving a high-profile role at Google so he could openly discuss his concerns.

Here’s some of what Hinton has said about the dangers of artificial intelligence:

Artificial intelligence technology could “amplify social injustice, erode social stability, and weaken our shared understanding of reality that is foundational to society.”

AI makes “it much easier for authoritarian governments to manipulate the electorate with fake news.” We need “legislation making it illegal to produce or share fake images or videos unless they are marked as fake.” (We already do this with currency.) But in the U.S., “one of the major political parties has tied its fate to the successful propagation of fake news.”

Combat robots powered by artificial intelligence could “allow rich countries to invade poor ones more easily.” And military systems that use AI to automatically set and kill targets could make wars impossible to control.

There is no one more qualified to testify about the ways artificial intelligence could be misused or get out of control. And of course Hinton is hardly the only expert warning about the threat of artificial intelligence — many of the world’s leading AI technologists are expressing concern.

As a society, we need to put strong guardrails on artificial intelligence and on the for-profit companies that are already locked in an AI arms race. But it’s no secret that elected officials are often too slow in keeping up with evolving technology — and that they are very much influenced by Big Tech.

That’s why it is essential for the American people to demand that Congress do more to grasp the risks of artificial intelligence — and to address those risks with comprehensive legislation and regulations that create guardrails to protect us — before it’s too late.

Tell Congress:

No less an expert than Nobel laureate Geoffrey Hinton — the “godfather of AI” — has been sounding the alarm about the profound risks artificial intelligence poses. Those risks are so serious, and the systems are evolving so rapidly — in part because major corporations are already locked in an AI arms race — that there isn’t time for the government to do its usual bumbling with new technology. The American people need you to fully grasp the threat of AI and to pass comprehensive legislation to protect us.”

president@citizen.org

wgbgator · Oct 15, 2024

I respect his concerns, but this kind of rhetoric mainly serves to fuel investment in AI and functions as a hype man. AI is a scam, making it some scary thing with endless possibilities for social control does a lot of work for the people who are making money from it. AI sucks, it mainly just generates social media and digital content slop. Its just crypto for slightly smarter people.

slocala · Oct 15, 2024

wgbgator said: ↑

I respect his concerns, but this kind of rhetoric mainly serves to fuel investment in AI and functions as a hype man. AI is a scam, making it some scary thing with endless possibilities for social control does a lot of work for the people who are making money from it. AI sucks, it mainly just generates social media and digital content slop. It’s just crypto for slightly smarter people.
Click to expand...

I have my concerns — not all actors will be profit motivated. Quantum computing + AI that can learn for itself can play out all the scenarios faster than a human and make an educated conclusion. The issue ultimately is runaway code and AI that is not controlled in a broad based digital ecosystem. Malicious instructions that are self evolving could lay dormant and wreak havoc disassociated from the deployment.

docspor · Oct 15, 2024

wgbgator said: ↑

I respect his concerns, but this kind of rhetoric mainly serves to fuel investment in AI and functions as a hype man. AI is a scam, making it some scary thing with endless possibilities for social control does a lot of work for the people who are making money from it. AI sucks, it mainly just generates social media and digital content slop. Its just crypto for slightly smarter people.
Click to expand...

my thoughts until Sat morning 2 weeks ago when I discovered notebooklm. I linked my buddies faculty webpage & it made a delightful & insightful podcast of his entire career in about 5 mins. Next morning I wrote a fake history of my real friends' band, saved it as a .pdf & drug it into notebooklm & it made a killer podcast in about 5 mins. The next week, we had a job candidate coming to present his research. I dropped his paper in notebooklm, it made a 13 min podcast. I took a walk on a beautiful day & listened.

docspor · Oct 15, 2024

wgbgator said: ↑

I respect his concerns, but this kind of rhetoric mainly serves to fuel investment in AI and functions as a hype man. AI is a scam, making it some scary thing with endless possibilities for social control does a lot of work for the people who are making money from it. AI sucks, it mainly just generates social media and digital content slop. Its just crypto for slightly smarter people.
Click to expand...

this is your post podcastified...glonuts was the band name...i did not change the title of the doc

https://notebooklm.google.com/notebook/48d3d0b7-e5ab-46a0-904f-1d0a155baf59/audio

gator_lawyer · Oct 15, 2024

Trickster said: ↑

AI makes “it much easier for authoritarian governments to manipulate the electorate with fake news.” We need “legislation making it illegal to produce or share fake images or videos unless they are marked as fake.” (We already do this with currency.) But in the U.S., “one of the major political parties has tied its fate to the successful propagation of fake news.”

Click to expand...

Unlikely to survive First Amendment scrutiny with the courts we have.

l_boy · Oct 15, 2024

gator_lawyer said: ↑

Unlikely to survive First Amendment scrutiny with the courts we have.
Click to expand...

You are probably right, but does artificially generated speech generated by a computer fall under the realm of rights of individual citizens?

gator_lawyer · Oct 15, 2024

l_boy said: ↑

You are probably right, but does artificially generated speech generated by a computer fall under the realm of rights of individual citizens?
Click to expand...

Somebody is making the video.

l_boy · Oct 15, 2024

docspor said: ↑

this is your post podcastified...glonuts was the band name...i did not change the title of the doc

https://notebooklm.google.com/notebook/48d3d0b7-e5ab-46a0-904f-1d0a155baf59/audio
Click to expand...

I am trying this thing, and it takes a really long time about 10 minutes so far. It just shows how cpu and energy intensive AI is.

wgbgator · Oct 16, 2024

l_boy said: ↑

I am trying this thing, and it takes a really long time about 10 minutes so far. It just shows how cpu and energy intensive AI is.
Click to expand...

This is the other issue … super heating the planet just so somebody can post an ai generated black nazi on twitter.

slocala · Oct 16, 2024

Right now data centers use about 5% of our US energy demands. It is projected to increase to 25-30%. Computing power energy demands cannot be achieved with traditional O&G only — “drill baby drill” is for the Luddites. Invest in small scale nuclear now or we lose the economic war of the future.

l_boy · Oct 16, 2024

wgbgator said: ↑

This is the other issue … super heating the planet just so somebody can post an ai generated black nazi on twitter.
Click to expand...

the tragedy is after about 15 minutes I accidently closed the browser, so I warmed the planet 0.01 degrees for no good reason. Alas, the AI podcast of the Too Hot Billy Napier thread would have been epic.

thomadm · Oct 16, 2024

gator_lawyer said: ↑

Somebody is making the video.
Click to expand...

Not really, its software generated content. Someone "owns" the data rights to the software AI, so there needs to be some laws in that arena to deal with it.

gator_lawyer · Oct 16, 2024

thomadm said: ↑

Not really, its software generated content. Someone "owns" the data rights to the software AI, so there needs to be some laws in that arena to deal with it.
Click to expand...

The software isn't generating itself. A person makes the requests and writes the prompts.

thomadm · Oct 16, 2024

gator_lawyer said: ↑

The software isn't generating itself. A person makes the requests and writes the prompts.
Click to expand...

Some of it, but a lot of software now can be generated by AI.

That's why I said someone needs to be able to own data rights. It's something that will be a huge problem in litigation.

WESGATORS · Oct 16, 2024

AI makes “it much easier for authoritarian governments to manipulate the electorate with fake news.” We need “legislation making it illegal to produce or share fake images or videos unless they are marked as fake.” (We already do this with currency.) But in the U.S., “one of the major political parties has tied its fate to the successful propagation of fake news.”
Click to expand...

Sounds simple, how would we begin to define "fake?" Now take that definition and apply it to artwork. What if somebody takes a real image and presents it as something else?

I went to the beach today...

This could both be true and misleading (the picture is the bottom of a car that needs to be repaired); maybe that's the car I took to the beach and I'm not trying to present the image as what I saw when looking out from the beach.

On second thought, maybe this would be the end of commercials or at least more honest ones:

Go GATORS!
,WESGATORS

mrhansduck · Oct 16, 2024

docspor said: ↑

this is your post podcastified...glonuts was the band name...i did not change the title of the doc

https://notebooklm.google.com/notebook/48d3d0b7-e5ab-46a0-904f-1d0a155baf59/audio
Click to expand...

I may be an outlier here, but I think this is very cool. Do you have to pay for it?

LimeyGator · Oct 16, 2024

Trickster said: ↑

“Last week, the Nobel Prize in Physics was awarded to two scientists — John J. Hopfield and Geoffrey E. Hinton — for research on computers that led to current AI technology like ChatGPT.

Hinton is often referred to as the “godfather” of artificial intelligence.

He is also outspoken about the serious risks AI poses to humanity — even leaving a high-profile role at Google so he could openly discuss his concerns.

Here’s some of what Hinton has said about the dangers of artificial intelligence:

Artificial intelligence technology could “amplify social injustice, erode social stability, and weaken our shared understanding of reality that is foundational to society.”

AI makes “it much easier for authoritarian governments to manipulate the electorate with fake news.” We need “legislation making it illegal to produce or share fake images or videos unless they are marked as fake.” (We already do this with currency.) But in the U.S., “one of the major political parties has tied its fate to the successful propagation of fake news.”

Combat robots powered by artificial intelligence could “allow rich countries to invade poor ones more easily.” And military systems that use AI to automatically set and kill targets could make wars impossible to control.

There is no one more qualified to testify about the ways artificial intelligence could be misused or get out of control. And of course Hinton is hardly the only expert warning about the threat of artificial intelligence — many of the world’s leading AI technologists are expressing concern.

As a society, we need to put strong guardrails on artificial intelligence and on the for-profit companies that are already locked in an AI arms race. But it’s no secret that elected officials are often too slow in keeping up with evolving technology — and that they are very much influenced by Big Tech.

That’s why it is essential for the American people to demand that Congress do more to grasp the risks of artificial intelligence — and to address those risks with comprehensive legislation and regulations that create guardrails to protect us — before it’s too late.

Tell Congress:

No less an expert than Nobel laureate Geoffrey Hinton — the “godfather of AI” — has been sounding the alarm about the profound risks artificial intelligence poses. Those risks are so serious, and the systems are evolving so rapidly — in part because major corporations are already locked in an AI arms race — that there isn’t time for the government to do its usual bumbling with new technology. The American people need you to fully grasp the threat of AI and to pass comprehensive legislation to protect us.

Last week, the Nobel Prize in Physics was awarded to two scientists — John J. Hopfield and Geoffrey E. Hinton — for research on computers that led to current AI technology like ChatGPT.

Hinton is often referred to as the “godfather” of artificial intelligence.

He is also outspoken about the serious risks AI poses to humanity — even leaving a high-profile role at Google so he could openly discuss his concerns.

Here’s some of what Hinton has said about the dangers of artificial intelligence:

Artificial intelligence technology could “amplify social injustice, erode social stability, and weaken our shared understanding of reality that is foundational to society.”

AI makes “it much easier for authoritarian governments to manipulate the electorate with fake news.” We need “legislation making it illegal to produce or share fake images or videos unless they are marked as fake.” (We already do this with currency.) But in the U.S., “one of the major political parties has tied its fate to the successful propagation of fake news.”

Combat robots powered by artificial intelligence could “allow rich countries to invade poor ones more easily.” And military systems that use AI to automatically set and kill targets could make wars impossible to control.

There is no one more qualified to testify about the ways artificial intelligence could be misused or get out of control. And of course Hinton is hardly the only expert warning about the threat of artificial intelligence — many of the world’s leading AI technologists are expressing concern.

As a society, we need to put strong guardrails on artificial intelligence and on the for-profit companies that are already locked in an AI arms race. But it’s no secret that elected officials are often too slow in keeping up with evolving technology — and that they are very much influenced by Big Tech.

That’s why it is essential for the American people to demand that Congress do more to grasp the risks of artificial intelligence — and to address those risks with comprehensive legislation and regulations that create guardrails to protect us — before it’s too late.

Tell Congress:

No less an expert than Nobel laureate Geoffrey Hinton — the “godfather of AI” — has been sounding the alarm about the profound risks artificial intelligence poses. Those risks are so serious, and the systems are evolving so rapidly — in part because major corporations are already locked in an AI arms race — that there isn’t time for the government to do its usual bumbling with new technology. The American people need you to fully grasp the threat of AI and to pass comprehensive legislation to protect us.”

president@citizen.org
Click to expand...

I used ChatGPT to summarise your post:

AI pioneer Geoffrey Hinton, a Nobel laureate, has raised serious concerns about the risks of AI, including its potential to amplify social injustice, manipulate elections, and escalate conflicts. He urges immediate action from Congress to regulate AI and protect society from its potential harms.

Not all bad.

docspor · Oct 16, 2024

l_boy said: ↑

the tragedy is after about 15 minutes I accidently closed the browser, so I warmed the planet 0.01 degrees for no good reason. Alas, the AI podcast of the Too Hot Billy Napier thread would have been epic.
Click to expand...

here it is. only took about 3 min. Have not heard it

https://notebooklm.google.com/notebook/e4ebcded-9176-4ff7-9ff5-e411eebd4c14/audio

edit. it is not very good

docspor · Oct 16, 2024

mrhansduck said: ↑

I may be an outlier here, but I think this is very cool. Do you have to pay for it?
Click to expand...

it's currently free.

Log in

Welcome home, fellow Gator.

The Gator Nation's oldest and most active insider community

Nobel Laureate’s AI warning

Trickster VIP Member

wgbgator Premium Member

slocala VIP Member

docspor GC Hall of Fame

docspor GC Hall of Fame

gator_lawyer VIP Member

l_boy 5500

gator_lawyer VIP Member

l_boy 5500

wgbgator Premium Member

slocala VIP Member

l_boy 5500

thomadm VIP Member

gator_lawyer VIP Member

thomadm VIP Member

WESGATORS Moderator VIP Member

mrhansduck GC Hall of Fame

LimeyGator Official Brexit Reporter!

docspor GC Hall of Fame

docspor GC Hall of Fame

Log in

Chat

Forums

Insiders Only

Free Sports

Plaza

Members

Welcome home, fellow Gator.

The Gator Nation's oldest and most active insider community

Nobel Laureate’s AI warning

Trickster VIP Member

wgbgator Premium Member

slocala VIP Member

docspor GC Hall of Fame

docspor GC Hall of Fame

gator_lawyer VIP Member

l_boy 5500

gator_lawyer VIP Member

l_boy 5500

wgbgator Premium Member

slocala VIP Member

l_boy 5500

thomadm VIP Member

gator_lawyer VIP Member

thomadm VIP Member

WESGATORS Moderator VIP Member

mrhansduck GC Hall of Fame

LimeyGator Official Brexit Reporter!

docspor GC Hall of Fame

docspor GC Hall of Fame