Crazy how the big yellow box is missing two words, that the bill is preventing the regulation of AI At the state level for 10 years.
I mean, why would AI be a states rights issue? Requiring that AI be regulated at the federal level makes a lot of sense when it comes to development, Sam Altman made a great point where trying to jump around a patchwork of regulation is really going to slow things down. Did Texas and Florida and a couple other states requiring ID verification make sites like Porn Hub include ID verification on their site, or did it just make Porn Hub block users in those states?
What if this picture of John Konrad was legal in my state, but illegal in New York?
I think we should let the technology mature for a while before we let state level lawmakers clip the wings of what can be a great tool.
When they overturned Roe vs Wade in favor of states rights, they opened the door for this kind of fragmented regulation
It’s at the very beginning of the meme:
“Subsection (c) states…”
Of course it should be regulated at the federal level. Right now, there’s no regulation at the federal level. I don’t see regulation happening at the federal level prior to this bill passing, and that seems to leave a can of worms opened, with the nation exposed to whatever regs this wealth-driven administration happens to strong-arm congress into passing. The ultra-wealthy tech industry, you can be sure, will get what they want.
“no state or political subdivision may enforce any law or regulation regulating artificial intelligence models, artificial intelligence systems, or automated decision systems.”
Meaning AI can not be regulated at the state or local level, but it does not prohibit it being regulated at the federal level.
The image is misleading, and a form of misinformation. It implies, (and I am under the impression you believe) the bill says “No one can regulate AI for 10 years” when the bill is actually saying “Only the federal government can regulate AI for 10 years.”
I understand that perfectly. What I’m stating is that, given the lack of actual federal regulation, there are no protections in place when this reconciliation bill passes. What then?
Presumably, Federal regulations… eventually. Trump already has an executive order to accelerate American AI development. Preventing State level legislation from fouling development, seems to be in line with that. The Technology is very new, and also AI is such a big umbrella, so until something with enough impact happens I’d imagine we won’t see regulation until it becomes obvious we need it.
There’s such a strong knee jerk reaction to AI, and I get it, it can be scary, but it has so much potential as a tool I don’t think it should be smothered.
Genuine question, what type of AI regulations would you want to see?
The big regulation I would like to see would be confining the use of AI to problem solving, and absolutely not allowing AI to make and enforce decisions that affect humans. I’ll have to think more about this.
Have You created this pic with AI research tool and AI pic. tools???
This is absolutely FUNeffingTASTIC !!! If it searched all available comments on the net and other sources and come up with this graphic idea it is absolutely great and FITTING.
I do recall using a fitting desription on this forum : His Majesty Captain John Konrad 
.
And tell You what : majestic handsomness and authority is emenating so strongly that i feel it thousands of miles away. WOW !!!
Listen . Now I am damn serious about this “devil incarnate” AI thing.
If it can do such thing like : instantaneous research of web and other source, graphic, voice manipulation/engineering of static pictures and video clips, can it then handle hand writting.
Example: You have a large stock of handwritten letters from your loved one -whomever( mother , father, wife etc, etc) collected during many years.
You feed them into a scanner collecting the database of hand written items
You feed it then into computer AI enhanced /aided program for hand written style analyses and create a " DEED AND /OR TESTAMENT" including signatures of atesting person including signature and hand written full name.
Next you run such created document through superduper high quality printer and hey presto . The document is ready.
Have You come across in your AI research about such AI aided program
The court expert will testify under oath the handwritting is genuine and same as delivered samples , so the only proof to the contrary can be to demonstrate the fake doc was not written by human but by the machine and this would require an indepth forensic analyses of the fake AI aided document.
What is your opinion??? .
I haven’t seen the image models handle text all that well, but forgeries have existed just about as long as paperwork has existed, so I’m sure that is probably on the way if not already here. It’s nothing photoshop probably couldn’t already handle, just lowering the barrier to entry.
But just today I was hearing about a model trained to convert the sound of a keyboard into text using AI.
I am more worried about video and audio AIs being more and more capable to copy someone’s likeness and used that to scam family member. That’s why my parents and I have a duress word just in case they get a call I was captured by the cartel or something, Imagine getting a facetime call from your kid asking for $5,000 but it’s not them.
There’s also the giant risk of some sort of combination of “boy who cried wolf” and large scale misinformation. Either people not believing something is happening because AI exists, or believing something that’s not happening because of AI. Is Admiral Konrad this much of a wild dancer? who knows.
Has anyone seen this bit of maritime related AI animation yet?
It shows an AI generated captain droning on about training. The script is an AI product cobbled together from forum posts somewhere, I’m guessing from an Indian maritime website.
I would think most people would find it convincing for about five seconds and then figure out what they were seeing was animation.
But then read the comments. Either the majority of viewers were fooled into thinking the animation was real, or alternatively, they accepted AI-generated talking points as valid as if a human had generated them.
The scary part is that we are at the same point in AI evolution as the Kodak Brownie camera was in 1900. From here the tech will get more convincing at a frightening pace.
Geez, the image alone is so obviously AI, it made me laugh. It’s bound to get better, though, and that’s a problem.
Edit: it’s what I would call the “smoothness” of the details
The “talking to the Camera” AI videos can actually be pretty convincing. If you know what to look for a lot of advertising and product endorsement is actually just AI.
I’m not entirely convinced this is a real video/person/profile:
Their content shifted from a hiking page to a far left soapbox and there is just something off about the video itself, and to be so polarizing, it smells fishy.
Also consider a 3rd option, it AI talking to AI. The problem is rampant on X (Formerly twitter) and Reddit, where so many accounts are just AI profiles, there’s a whole conspiracy about the Dead Internet theory.