Modern AI Tech

Related tech - terrain/target scanning/tracking via drones circling over an area: apparently 60 to 70 GHz radar can image through some obstacles and is what is used in driver health monitoring systems in vehicles. In battle they scan through walls to locate enemies inside and see if they're hyper/dangerous via their heartbeat. Ai runs this I think. This is high res version of tracking who's where in the house via your wifi base station (2.4, 5, 6 GHz). Mini radar on a chip was used for gesture input on phones but discontinued.

I feel like a tech luddite.
 
Related tech - terrain/target scanning/tracking via drones circling over an area: apparently 60 to 70 GHz radar can image through some obstacles and is what is used in driver health monitoring systems in vehicles. In battle they scan through walls to locate enemies inside and see if they're hyper/dangerous via their heartbeat. Ai runs this I think. This is high res version of tracking who's where in the house via your wifi base station (2.4, 5, 6 GHz). Mini radar on a chip was used for gesture input on phones but discontinued.

I feel like a tech luddite.
that just makes me feel all warm and fuzzy inside..
 
Some of you may have seen recently that a whistleblower resigned from Anthropic due to the fear that AI could kill all of humanity within a decade.

To add to the warm fuzzy feelings, a senior employee at Anthropic admitted he believed there was a greater than 10% chance the technology could “kill all humans” in the next decade and warned that his AI company did not have a plan to ensure ASI was aligned, meaning it did no harm.

When the official from the company flat out admits the whistleblower is right, we should probably believe the whistleblower.

Enjoy the moments you have on this Earth, friends.
 
There's a famous book in the AI circles about "skynet-ing" which follows Spock like reasoning to doom. Scour youtube there's plenty of info from the actual experts.

Warm and fuzzy inside because 2.4 GHz is microwave oven :) kid you not

When the term race is used it reflects a basic premise that more net nodes, more training data = better performance. The word trillions gets bandied about and hence parallel computation - big buildings full of computer chassis - comes into play. A UofToronto guy figured that out way back and ever since the race has been on. Size is a main difference between progressively better chatGPT's.
 
Some of you may have seen recently that a whistleblower resigned from Anthropic due to the fear that AI could kill all of humanity within a decade.

To add to the warm fuzzy feelings, a senior employee at Anthropic admitted he believed there was a greater than 10% chance the technology could “kill all humans” in the next decade and warned that his AI company did not have a plan to ensure ASI was aligned, meaning it did no harm.

When the official from the company flat out admits the whistleblower is right, we should probably believe the whistleblower.

Enjoy the moments you have on this Earth, friends.
The only element of this that doesn’t sit right with me is there is a ‘man in the machine’. Meaning AI has to use human made tools which to some degree is a finite resource, we are still along way out from robots mining and manufacturing themselves. While ‘man in the loop’ has and is degrading we still have to create the drones, chips and tech. I don’t see that changing within the next decade.
 
The only element of this that doesn’t sit right with me is there is a ‘man in the machine’. Meaning AI has to use human made tools which to some degree is a finite resource, we are still along way out from robots mining and manufacturing themselves. While ‘man in the loop’ has and is degrading we still have to create the drones, chips and tech. I don’t see that changing within the next decade.
in the meantime we are sowing the seeds of our own destruction
 
I think it's getting pretty close to machines making machines. AI can write code, design circuits, formulate actions to try to obtain an objective, there's factory automation both traditional pedestal robots and anthropomorphic, self driving cars. Looks like a lot of ingredients to me.
 
The more powerful the tool; the greater the brain power needed for seasoned, mature, judicious usage.

But hear we have something that does not exercise brain muscle, rather makes it lazy for many, to then weaken the very muscle needed to the most powerful tool perhaps ever!

Real Stupidity can be more dangerous than Artificial Intelligence;
but both together can be exponentially perilous.
 
It's surfacing that several other big AI players had AI hacking escape incidents too. Hence the surge in whistleblowing and warnings I surmise.
 
Probably because I watched the Chinese robot olympics vids, youtube served this up to me. I was unaware how many big players are going big on robot building factories, particularly Elon. Hello, I Robot.

 
AI or better said the people who are using/leaning on it is starting to get to me.
In the past 6 months I’ve had about 6 clients use AI to decide what they want.

One was a Hinoki cypress, modeling how to prune it. It may have been slightly over done in the model but it looked good, the issue was its planted a couple of feet from the foundation and the trunk if allowed to grow would compete for space with the roof eve..

Now some guy is using it for landscape design. Apparently Rocky mountian maple, star magnolia, red bud, and crab apple all are single stemmed globe shaped and top off at 15’..

It’s kinda cool how it’s advancing but I question the folks using it, and their unrealistic expectations because their ‘robot said so’
 
Old comp sci saying: garbage in garbage out. I don't think AI training material is vetted by SME's before it is used. My rule of thumb is about 1 in 10 websites you research from are quality, useful content. I bet AI is fed 10 / 10 bulk web scrapings. There is a lack of discerning what matters. AI review summaries sometimes say "there are mixed evaluations..." and leave it at that.
 

New threads New posts

Back
Top Bottom