john12ax7
Well-known member
The people I know who have been in the room with him don't seem to share that opinion.
Dangers of AI
- Thread starter boji
- Start date
Help Support GroupDIY Audio Forum:
The people I know who have been in the room with him don't seem to share that opinion.
ruffrecords
Well-known member
Cadence just announced its latest PCB layout package with AI routing. Claims to improve PCB turnaround by 10 times.
Cheers
Ian
Cheers
Ian
I saw this video recently. But I'm sure if Cadence decides to implement something similar it will work more reliably. Or their marketing department is just playing buzzword bingo.
Last edited:
I know a person who worked as an interpreter most of their career ,
A computer program does the basic translation , but what the machine spits out is often garbage until the interpreter goes back over and corrects things ,
Seems like the versions of chat AI were seeing now are similar ,
Then again imagine a drone aircraft , it could quite easily be sent off on an autonomous search and destroy mission , electronic eaves dropping , camera , biometrics and in for the kill .
All the AI needs is a small number of co-opperative humans to wipe its arse for it everyonce in a while.
Theres a buddy of mine works in the UK in a place dedicated to military grade AI systems ,
I must try and catch up with him next time he's back ,
even 10-15 years ago the stuff he was suggesting seemed much more advanced than the latest crop of computer augmented learning systems appearing now .
Imagine a glorified chess computer , but instead of pieces on a board it moves a cavalry division or calls in air support , the only reason humans are even there is to press the reset button if the AI has a brainfart .
A computer program does the basic translation , but what the machine spits out is often garbage until the interpreter goes back over and corrects things ,
Seems like the versions of chat AI were seeing now are similar ,
Then again imagine a drone aircraft , it could quite easily be sent off on an autonomous search and destroy mission , electronic eaves dropping , camera , biometrics and in for the kill .
All the AI needs is a small number of co-opperative humans to wipe its arse for it everyonce in a while.
Theres a buddy of mine works in the UK in a place dedicated to military grade AI systems ,
I must try and catch up with him next time he's back ,
even 10-15 years ago the stuff he was suggesting seemed much more advanced than the latest crop of computer augmented learning systems appearing now .
Imagine a glorified chess computer , but instead of pieces on a board it moves a cavalry division or calls in air support , the only reason humans are even there is to press the reset button if the AI has a brainfart .
Then why is he running Twitter into the ground, haphazard shadow banning anything he doesn't like?
Current active Twitter users: between 100.000 and 200.000 people who post at least once a month. And that number is dwindling.
Commercial bots are estimated to at least 1.000.000. Political bots even more. Nothing has been done, as the commercial bots belong to advertisers.
I've heard from someone who was fired, the moderation team is so small now they can hardly handle mail, let alone what nefarious bots are doing.
I don't think Twitter's data stream is even useful for AI. Far tto much noise and little signal. Reddit is worse and someone already has an AI that behaves like 4chan...
Besides, if he's so smart, why did he retract from Open AI in the first place? In need of money?
That was a monumental mistake.
john12ax7
Well-known member
The problem with things like auto routers has often been the time it takes to review and correct the routing can rival just doing it the right way from the start. Maybe AI will fix that.
I imagine it's similar in lots of fields. But what happens when the corporate bean counters decide AI is good enough so that human QC is eliminated altogether? One potential preemptive solution would be to eliminate or reduce the corporate shield for personal responsibility, that would certainly reduce reckless behavior.
I imagine it's similar in lots of fields. But what happens when the corporate bean counters decide AI is good enough so that human QC is eliminated altogether? One potential preemptive solution would be to eliminate or reduce the corporate shield for personal responsibility, that would certainly reduce reckless behavior.
![Electronics Soldering Iron Kit, [Upgraded] Soldering Iron 110V 90W LCD Digital Portable Soldering Kit 180-480℃(356-896℉), Welding Tool with ON/OFF Switch, Auto-sleep, Thermostatic Design](https://m.media-amazon.com/images/I/41gRDnlyfJS._SL500_.jpg)
PermO
Well-known member
The gpt chatbot is really helpfull when you don't know how to write code for an Arduino, it spits it out in no time.
Today I had lunch with a software engineer and told him about that, he never used chatgpt so he did a little testrun asking it to write some code, just some basic stuff, but it was flawless and superfast.
He was impressed and will defenitely try it on some more advanced coding.
Today I had lunch with a software engineer and told him about that, he never used chatgpt so he did a little testrun asking it to write some code, just some basic stuff, but it was flawless and superfast.
He was impressed and will defenitely try it on some more advanced coding.
he is surely smarter than me... Opinions vary but purchasing twitter was not his wisest move.
He just revealed that the government was able to read twitter user's private messages.
His characterization of Twitter was that it was an activist organization. He seems to be running that politcal activism into the ground.many people are resisting paying for services.
I would not expect a friendly response from somebody who was fired.
He says he was squeezed out. He recognized the risk from AI day one, and that position was not appreciated by the other investors.
he has made more mistakes than people who do nothing but criticize, but he has done some remarkable stuff also.
JR
AnalogPackrat
Well-known member
References?
He admitted in a recent interview that he wasn't sure he could make it financially viable. When asked if he regretted buying it said he didn't. Uncovering the collaboration between big tech and gov was important enough to offset the cost.
In the same interview Musk stated that they'd drastically reduced bots and fake account bad actors. Where does your "estimate" originate?
That contradicts what Musk said about the current employee headcount and operational efficiency/capability at Twitter.
You could say the same of much internet traffic as a whole. There's still some wheat amongst the chaff.
That was also asked in the interview. He regrets "taking his eye off the ball" with OpenAI, but also related the story of how, and more importantly, why, he founded OpenAI in the first place. This then led into why he's trying to get back into AI development again and to what end.
And, unlike most tech entrepreneurs, he openly admits these failures and mistakes.
I am not as good at mind reading as some here, but Elon tends to say what he is thinking often.
My speculation is that he has some larger future plan that involves using twitter as a platform for some of his other future technology ideas. I don't expect hime to make a working (profitable) version of the old twitter, but some new completely different business model.
Of course this is mostly guesswork, and the future has not happened yet so we'll see, or not.
JR
My speculation is that he has some larger future plan that involves using twitter as a platform for some of his other future technology ideas. I don't expect hime to make a working (profitable) version of the old twitter, but some new completely different business model.
Of course this is mostly guesswork, and the future has not happened yet so we'll see, or not.
JR
Musk didn't start OpenAI. He was just one of the investors.
From Wikipedia:
https://en.wikipedia.org/wiki/OpenAI#History
The info came from one of the former Twitter directors. He tweeted it, to have his account immediately ghost-banned. As if he wouldn't notice. So much for openness.
And yes, he's still selling the fairytale that he's one of the founders...
He also likes to lie about headcount: it's currently around 1.500 people, but Musk tells interviewers Twitter has a 2.300 headcount. And less than 550 are engineers. Nearly all the engineers in security and coding are gone. And most of the contractors don't feel like working for egomaniac Musk.
Do a bit of research and you'll see what's happening. Unless there's an injection of capital Twitter will be gone by the end of next year.
Have a look here:
https://www.businessinsider.com/twitter-revenue-per-employee-collapsed-before-musk-takeover-2023-4
And this the biggest problem:
https://www.businessinsider.com/twitter-sign-up-searches-plunged-81-since-elon-musk-takeover-2023-4
No new users...
From Wikipedia:
https://en.wikipedia.org/wiki/OpenAI#History
The info came from one of the former Twitter directors. He tweeted it, to have his account immediately ghost-banned. As if he wouldn't notice. So much for openness.
And yes, he's still selling the fairytale that he's one of the founders...
He also likes to lie about headcount: it's currently around 1.500 people, but Musk tells interviewers Twitter has a 2.300 headcount. And less than 550 are engineers. Nearly all the engineers in security and coding are gone. And most of the contractors don't feel like working for egomaniac Musk.
Do a bit of research and you'll see what's happening. Unless there's an injection of capital Twitter will be gone by the end of next year.
Have a look here:
https://www.businessinsider.com/twitter-revenue-per-employee-collapsed-before-musk-takeover-2023-4
And this the biggest problem:
https://www.businessinsider.com/twitter-sign-up-searches-plunged-81-since-elon-musk-takeover-2023-4
No new users...
john12ax7
Well-known member
Much like with Tesla. Musk is what happens when above average intelligence combines with an extreme grandiose sense of self and no moral compass. You can go quite far in business and finance with this. As Munger says "Don't underestimate the man who overestimates himself". But extreme financial success is not the same as extreme intelligence.
AnalogPackrat
Well-known member
Read your wikipedia link more carefully. He's listed as one of the founders. Founders can also bring initial capital investment which likely several did. I never said he was the sole founder. In his relation of the story he and one other person were the ones who wanted to do "safe AI" after he had a conversation with Larry Page who stated he wanted to create an AI god or somesuch and was not concerned about consequences.
Never heard about it. Given the deeply sinister crap the company was doing before I wouldn't trust anything from its former "leaders" who clearly lied about all kinds of things in the recent past.
Of what? OpenAI? See above.
It's a privately held company. Where's the source of this info? Also, operations seem to be running despite rumors to the contrary. I worked in Silicon Valley from 1993-2021 and lived through several boom-bust cycles. I've worked for a few grossly inefficient companies with huge excess staff. One place folded, others were acquired (one fire sale) followed by mass layoffs, and one is still floundering while burning huge amounts of $$$ on stuff that won't payoff. There are plenty of egos in SV at all levels. Musk is far more successful than most.
Yes, Twitter was horribly inefficient before he bought it. He made it better.
According to some derivative measure by an anti free speech Google. Time will tell. It's his money and his company now whether you like it or not.
It seems AI might be a better solution for the IRS instead of 87k new agents. Reportedly the IRS technology is archaic.
JR
JR
AnalogPackrat
Well-known member
A flat tax would solve a lot of problems and reduce the need for audits, etc. allowing the IRS to shrink rather than grow. But there's a huge industry dependent on keeping taxes complicated, so good luck.
if anything they want to make it even more progressive....
Using AI instead of hiring new human agents is at least possible. AI would probably be more competent.
JR
AnalogPackrat
Well-known member
You can't fix stupid, apparently.
It would only be fair if we can use AI to do our taxes to start with.
AnalogPackrat
Well-known member
He seems to have a more functional moral compass than many (e.g., Steve Jobs, Bill Gates, Brin and Page, Vidjaya Gadde, Zuckerberg, etc.). Where would Tesla be if he hadn't taken control? Note: I'm no fan of Tesla.
Sure, but it takes a lot more than that to succeed in multiple tech domains like he has, from cryptography and SW at PayPal to battery, electric motor, controls, advanced SW, and mass production at Tesla to rockets, fuels, and more advanced SW at SpaceX (where he's undercut NASA launch costs by at least an order of magnitude and taken other competitors' business as well). He's not just some well-connected financier or VC asshat.
Sounds like sour grapes with a tinge of envy or jealousy. Have you or your "friends in the room" with Musk built any companies comparable to SpaceX? Even PayPal?
not only can't you fix it, we can't escape it.
Turbo tax is getting pretty close.... (first year I declared that I am "retired"
).JR
PS: I don't think Elon bought Twitter for moral reasons, but it isn't difficult to out point that group. BTW I never met a successful businessman with low self esteem.
boji
Well-known member
Not sure if this was replied to elsewhere above, but I'm fairly sure the reason he bailed on OpenAI was because they (⸮⸮⸮) took the IP private.
Last edited:
