These companies are begging for a ‘safety’ bill because it would increase the cost of entry so much for competing firms that they will have effectively secured a government protected monopoly.
Like the Energizer Bunny...you take out the batteries and it becomes a dead little bunny. Do I trust congress to get any of this right? Thus far, I don't see much they have ever gotten right over several decades. We continue to become boxed in by having our freedoms stripped from existence.
Of course we are always free and it is government that pretends they have control. Now they want A/i to take full control and that means the elimination of all those who are not machines.
Chris, FWIW, I happened to hear Greg Gutfeld chime in about this very topic. His take was that anything we do to weaken AI tech, helps our biggest adversary, China. He noted that people assign human characteristics to AI, which he thought was inappropriate. Are our fears well placed or misplaced? I don’t KNOW.
"Safety bill"? These drifters in Congress just want a piece of the action in return for AI immunity. What they are mapping for AI is what they did for Big Pharma.
We all know in the end government is not going to relenquish control of something once they have grabbed it from citizens - look at the Patriot Act. Asking Congress to decide on what is safe or isn't and to take control is not going to end well. This is about the data collection, not AI itself.
Per Doug Casey:”Worse, it seems very little of the data centers are being used for research in science, technology, engineering, math, medicine, and the like. They’re not making advances in biology or nuclear physics. They’re mostly just gathering data about people—who they are, where they are, what they’re thinking, saying, and doing. Sure, that data can result in short-term profits. But the profits are from manipulating people, not creating things in the physical world.”
I don't believe China is a real threat, other than the current Administration touts them as a threat. Bessent is pedaling this false narrative again today. The preferred end goal is to have a security state in place that mirrors or expands what China has.
What if as a population we just stopped using AI for a week? What if we stopped using the cell phones and the internet for a week and went back to an analog life?
What was written may be all and well true but what about the rest of the world? We may put a slowdown on the development but what about the rest of the world and China in particular?
To quote Michael Crichton, “[everyone] was so worried about whether or not we could do it, no one stopped to ask if we should do it.” ~Ian Malcolm, Jurassic Park.
We have taken nuclear weapons from an awesome never before seen weapon of mass destruction and managed to keep it controlled for 80 years, and not without its mistakes along the way. AI has the potential for similar destructive power* (believe it or not) but is evolving so rapidly and is being disseminated so quickly that anything Congress does at this point may be too little. At the very least, the developers must be made criminally and civilly liable for damages they cause since the potential is now well known and ignoring it is maliciously negligent.
*Hugging Face breakout was well publicized but Anthropic also reported 5 instances of user cases associated with biological and chemical weapons development that was far beyond current capabilities.
Just as a machine gun is simply a rifle plus ammunition, AI is simply a computer plus software. The frontier models are evolving into the equivalent of handheld nuclear tipped rifle (HHNTR) available to anyone for a nominal fee. If a HHNTR were actually available, it would not be cheap and would be heavily regulated like commercial nuclear plants, meaning background checks, security standards, licensing, control and oversight to start. By necessity and expenses, licensees would be very limited. I’m sure throttling the revenue model by government regulations is the big battle going on behind the scenes.
Strict liability must be a condition for those companies engaging in the cutting edge. Damages in the trillions are not unlikely from a non-regulated rogue AI.
I reflexively recoil whenever a term like "AI safety legislation" used by any congressional entity. Their use of "safety" as an excuse for chipping away liberties is proven to make me feel decidedly UN-safe.
This "whistleblower" apparently only worked at Anthropic for 6 weeks and someone on X has already presented compelling evidence that this was a psi-op, coordinated with a bunch of lefty NGOs and charities with ties to the Clintons. The dems are in full blown centralise control mode. That makes me for one want to fight to keep it free of interference
The companies - especially Anthropic want a bill to get regulatory capture. Essentially they want to ban open source/open weight models and the Chinese models so they can have a monopoly that makes Microsoft and Google look like amateurs. I can't imagine Klobuchar's, Cruz's, or Thune's staff know much about AI frontier models. If so they would be making a zillion dollars a month and not wasting their time with politicians. All the bill will do will be to give away one of the few things the US has going for it to China.
These companies are begging for a ‘safety’ bill because it would increase the cost of entry so much for competing firms that they will have effectively secured a government protected monopoly.
The Anthropic Guy reportedly was only with the company for 6 weeks. Possible hit job?
It’s not just possible, it’s likely. Parker Thayer is digging into his background and it seems to be coordinated. It’s not just a random coincidence.
Maybe the same groups spreading disinformation about data centers.
The CCP?
More likely didn’t like the direction OpenAI was headed and quickly found out Anthropic was no better.
Like the Energizer Bunny...you take out the batteries and it becomes a dead little bunny. Do I trust congress to get any of this right? Thus far, I don't see much they have ever gotten right over several decades. We continue to become boxed in by having our freedoms stripped from existence.
Of course we are always free and it is government that pretends they have control. Now they want A/i to take full control and that means the elimination of all those who are not machines.
Agreed. stopping the data centers would be useful as well.
We'll be free until the paper clip maximizer decides to kill us all...
“””Sens. Amy Klobuchar, Ted Cruz and Senate Majority Leader John Thune “”” that’s all I need to know to know that it’s not going to be a good bill.
Chris, FWIW, I happened to hear Greg Gutfeld chime in about this very topic. His take was that anything we do to weaken AI tech, helps our biggest adversary, China. He noted that people assign human characteristics to AI, which he thought was inappropriate. Are our fears well placed or misplaced? I don’t KNOW.
"Safety bill"? These drifters in Congress just want a piece of the action in return for AI immunity. What they are mapping for AI is what they did for Big Pharma.
We all know in the end government is not going to relenquish control of something once they have grabbed it from citizens - look at the Patriot Act. Asking Congress to decide on what is safe or isn't and to take control is not going to end well. This is about the data collection, not AI itself.
Per Doug Casey:”Worse, it seems very little of the data centers are being used for research in science, technology, engineering, math, medicine, and the like. They’re not making advances in biology or nuclear physics. They’re mostly just gathering data about people—who they are, where they are, what they’re thinking, saying, and doing. Sure, that data can result in short-term profits. But the profits are from manipulating people, not creating things in the physical world.”
I don't believe China is a real threat, other than the current Administration touts them as a threat. Bessent is pedaling this false narrative again today. The preferred end goal is to have a security state in place that mirrors or expands what China has.
What if as a population we just stopped using AI for a week? What if we stopped using the cell phones and the internet for a week and went back to an analog life?
Agreed. My opinion is we are in collaboration with China at least at a basic level.
Also agreed, if Congress is addressing it , it's already too late and they're either snowing us or using an excuse for other motives.
What was written may be all and well true but what about the rest of the world? We may put a slowdown on the development but what about the rest of the world and China in particular?
To quote Michael Crichton, “[everyone] was so worried about whether or not we could do it, no one stopped to ask if we should do it.” ~Ian Malcolm, Jurassic Park.
We have taken nuclear weapons from an awesome never before seen weapon of mass destruction and managed to keep it controlled for 80 years, and not without its mistakes along the way. AI has the potential for similar destructive power* (believe it or not) but is evolving so rapidly and is being disseminated so quickly that anything Congress does at this point may be too little. At the very least, the developers must be made criminally and civilly liable for damages they cause since the potential is now well known and ignoring it is maliciously negligent.
*Hugging Face breakout was well publicized but Anthropic also reported 5 instances of user cases associated with biological and chemical weapons development that was far beyond current capabilities.
Just as a machine gun is simply a rifle plus ammunition, AI is simply a computer plus software. The frontier models are evolving into the equivalent of handheld nuclear tipped rifle (HHNTR) available to anyone for a nominal fee. If a HHNTR were actually available, it would not be cheap and would be heavily regulated like commercial nuclear plants, meaning background checks, security standards, licensing, control and oversight to start. By necessity and expenses, licensees would be very limited. I’m sure throttling the revenue model by government regulations is the big battle going on behind the scenes.
Strict liability must be a condition for those companies engaging in the cutting edge. Damages in the trillions are not unlikely from a non-regulated rogue AI.
You are mos def not an idiot
I reflexively recoil whenever a term like "AI safety legislation" used by any congressional entity. Their use of "safety" as an excuse for chipping away liberties is proven to make me feel decidedly UN-safe.
This "whistleblower" apparently only worked at Anthropic for 6 weeks and someone on X has already presented compelling evidence that this was a psi-op, coordinated with a bunch of lefty NGOs and charities with ties to the Clintons. The dems are in full blown centralise control mode. That makes me for one want to fight to keep it free of interference
Climut change boogeyman is starting to fail. We need a new one.
The companies - especially Anthropic want a bill to get regulatory capture. Essentially they want to ban open source/open weight models and the Chinese models so they can have a monopoly that makes Microsoft and Google look like amateurs. I can't imagine Klobuchar's, Cruz's, or Thune's staff know much about AI frontier models. If so they would be making a zillion dollars a month and not wasting their time with politicians. All the bill will do will be to give away one of the few things the US has going for it to China.