Can we align human interests with robots, so they don’t turn on us?

Topic

New Reply

PUBLIC DEFENDER By Brian Livingston Robots in human-like forms are already starting to assume jobs that have been performed for centuries by ordinary
[See the full post at: Can we align human interests with robots, so they don’t turn on us?]

9 users thanked author for this post.

ibe98765, windbg, Myst, Lars220, fisher75, Sky, Izzienoel, Alex5723, DLivesInTexas

Reply | Quote

Replies

“This makes it more important than ever that we get an overriding moral code built into every intelligent robot and artificial-intelligence app right now.”

First, we need to figure out how to build basic moral codes into humans, especially candidates for political offices.

9 users thanked author for this post.

Toa Of Justice, Steve, ibe98765, Myst, samak, Charlie, WSbobby-p, Mr. Austin, NaNoNyMouse

Reply | Quote

Aye, there’s the rub 😉

Human, who sports only naturally-occurring DNA ~ oneironaut ~ broadcaster

Reply | Quote

Has someone told the armed forces of the world, and the industry that supports them, about the 1st law?  These will be the bots that destroy humanity…

2 users thanked author for this post.

Toa Of Justice, Mr. Austin

Reply | Quote

Hhhmm… maybe no one mentioned it to them.

Human, who sports only naturally-occurring DNA ~ oneironaut ~ broadcaster

Reply | Quote

Mr. @brian, Your Dudeness:

I love your thought and writing. You also have the option to look into, and get hip about Graham Hancock’s longstanding work and books. And if any of that resonates with you, please LMK, as there are other different researchers with facts which help triangulate Mr. Hancock’s. A few such authors are personal friends. The Younger Dryas comet might be an interesting place for you to start.

Human, who sports only naturally-occurring DNA ~ oneironaut ~ broadcaster

1 user thanked author for this post.

Lars220

Reply | Quote

Thank you Brian Livingston for this historic and educational article. It is very informative and entertaining. Well done sir! Here I will share a link to an article from World Bank Blog from year 2016:

Robots: What can workers do to protect themselves from automation?
by Harry A. Patrinos  –  January 11, 2016  –  World Bank Blogs

“Stephen Hawking has warned that artificial intelligence could end the human race. The development of intelligent machines could pose a major threat to humanity. Sometime in the near future, machine intelligence is predicted to surpass human intelligence, a point in time known as “the singularity.” Whether the rise of the machines is an existential threat to mankind or not, there is a more mundane issue—robotics are being used to automate production. There are more than 300,000 industrial robots in operation in Japan and another 200,000 in North America. This is seen by some as a threat to jobs.”

Amazing Truths Revealed: Link to Simply Magnificent AskWoody Knowledge Bases

3 users thanked author for this post.

SueW, Charlie, Myst

Reply | Quote

“To oversimplify here, the mechanical entities’ brains would be hard-coded to never injure humans and to always obey human orders.”

The big problem with Asimov’s laws are that they can only exist if the robots aren’t allowed to repair/modify each other, which is not likely.

The reality is that robots/AI will be controlling manufacturing and modification of their own images and therefore would be able to override any control imposed, whether software or hardware.

Humans (meat) will not be a protected class under the coming AI/robot oligarchy.

1 user thanked author for this post.

Charlie

Reply | Quote

“In the novelette With Folded Hands by Jack Williamson, robots took the phrase literally. They prevented humans from experiencing stress, unhealthy lifestyles, extreme sports, and anything else that could potentially harm a person. Humans could do nothing but sit with folded hands.”

This novelette was expanded into a couple of books called the Humanoids (1949) and a sequel, The Humanoid Touch (1980).  The robots exists to “To Serve and Obey, And Guard Men From Harm”.  It’s amazing visionary SF writing, of which there was much n early half of the 29th century.  Highly recommended!

https://en.wikipedia.org/wiki/The_Humanoids

Reply | Quote

Much SF has been written over futures where robots/AI’s and humans live together (or not).  Sometimes machines will be subservient to humans, more often they will not.

For those who like space opera with plenty of space battles, I’d recommend SF author Neal Asher, who has created a universe called the Polity that is also ruled by sentient machines that work with and against humans.  There are probably 30 odd books in this series (always start from book #1.  Look on Wikipedia for the book order)..

In this universe, the machines decided among themselves to take over when we humans started yet another war among ourselves but the machines that were supposed to fight refused.  Instead they assumed control of all government and military functions (after all, the humans were dependent on the machines for everything anyway).

There is a big honking AI in charge of everything that resides on Earth and a lot of minor AI robots moving throughout the galaxy that help do its bidding across the patch of the universe that it controls. There are apparently no politicians (surely a good thing!).

But AIs in this universe can go rogue (e.g. AI war machine Penny Royal) and create all sorts of mayhem, death and destruction, which spices things up.

The Polity is raw place where money, crime, various bad aliens and regular warfare still exist.  But it is ruled by an AI.

1 user thanked author for this post.

Lars220

Reply | Quote

Figure 02 Robot Is a Sleeker, Smarter Humanoid
By Evan Ackerman  –  August 6, 2024  –  IEEE Spectrum

“Today, Figure is introducing the newest, slimmest, shiniest, and least creatively named next generation of its humanoid robot: Figure 02. According to the press release, Figure 02 is the result of “a ground-up hardware and software redesign” and is “the highest performing humanoid robot,” – – – This thing looks slick. I’d say that it’s maybe a little too far on the sinister side for a robot intended to work around humans, but the industrial design is badass and the packaging is excellent, … “

Introducing Figure 02  –  YouTube  –  Aug 6, 2024  – 2 minutes
https://www.youtube.com/watch?v=0SRVJaOg9Co

Amazing Truths Revealed: Link to Simply Magnificent AskWoody Knowledge Bases

1 user thanked author for this post.

ibe98765

Reply | Quote

We haven’t been successful in aligning computers, the internet and social media with ideal human interests and values. Otherwise we wouldn’t have:  Viruses, Malware, Identity Theft, Ransomware, Dark Patterns, Surveillance, Misinformation, Propaganda, etc., etc. Different people use technologies for different ends.

We haven’t been able to “align” most every technology humans have developed. Technologies are used by ‘good’ people to do good things…. and bad things by ‘bad’ people. Why do techno-optimists think we can do it with AGI and robots?  The genie is already out of the bottle. AI is already out in the wild. AGI will likely take a similar path.

Besides, who actually gets to decide what “ideal human interests and values” actually means? And who will enforce them? Seems it’s us humans that need to “align” ourselves first. The age-old ethical problem of human nature, eh?

Win10 Pro x64 22H2, Win10 Home 22H2, Linux Mint + a cat with 'tortitude'.

2 users thanked author for this post.

Steve, Toa Of Justice

Reply | Quote

https://bgr.com/tech/robots-are-using-ai-to-train-themselves-which-isnt-at-all-terrifying/

Researchers at MIT have given robots everything they need to take over the world. A new algorithm called “Estimate, Extrapolate, and Situate (EES)” will soon allow robots to train themselves without human assistance…

The EES algorithm not only lets the robots train themselves, but it will also allow them to identify weaknesses in their skills. They can then take those weaknesses and use them to determine where they need to improve their skills…

The robot can then use the EES to determine if additional practice is needed to enhance the robot’s performance. If it is, then the robot self-training algorithm will create training material for the robot and put it to use…

Practice Makes Perfect: Planning toLearn Skill Parameter Policies

Reply | Quote

Let me tell you how the USA might get invaded quietly and bloodless by a foreign power using modern technology .

Washington is upset over the TikTok app because of fears that China is collecting info from USA users, could access data on their phones or could introduce malware onto their phones and then into connected networks, via the app.  Politicians are trying to ban the app or force its sale to a US business or owner.

But here’s another vector that China or any other state/government could use to infiltrate a country.

China, for example, is making big advances in the development of humanoid robots and some companies are deploying early versions now.  Those robots will be designed in a foreign country, could have back doors installed in them, will be programmed by a foreign owned business and likely, will be connected into networks and computers at companies they have been sold to.  They will be able to collect all sorts of proprietary and operational info from within companies, possibly over the networks they are connected, depending on firewall settings,  but also visually (by cameras).

This collected corporate intelligence could be transmitted home or if blocked by networks, could possibly be downloaded via say SD Card or by transfer to a plugged in flash drive by an employed agent.  Another possibility is that these robots could be used to transfer malware onto corporate and government networks.

I’ve not read any discussions on this huge security problem.  I hope people smarter than me are considering these potential security issues.

Reply | Quote

ibe98765 wrote:

This collected corporate intelligence could be transmitted home or if blocked by networks, could possibly be downloaded via say SD Card or by transfer to a plugged in flash drive by an employed agent.  Another possibility is that these robots could be used to transfer malware onto corporate and government networks.

All this is done today by the PCs/Smartphones.. you use. No need for robots.

Reply | Quote

“All this is done today by the PCs/Smartphones.. you use. No need for robots.”

Not true.  Off the top of my head, humanoid robots will be integrated into all aspects of company operations, seeing things that a cellphone in your pocket doesn’t see.  For example, counting the number and type of product boxes that a company is shipping as it loads them into trucks and possibly, where those boxes are going.  That’s valuable competitive info that could be sold to others or used in blackmail.

Reply | Quote