Artificial Intelligence - The Greatest Threat

[Bob Kuczewski]

Purpose

This Blog is dedicated to understanding the threat of Artificial Intelligence to humans.

History

On April 22, 2019, Bob Kuczewski quoted "A small voice in the future" as saying:

Tell me again Granpa ...
Tell me again what it was like when people were smarter than computers.

Rules

All members of the U.S. Hawks are welcome to post. The author reserves the right to remove posts that stray from the subject. Please try to be reasonably polite.

Background

I began working in "AI" in 1985. At that time we were struggling to solve even the simplest problems with machine learning. A lot has changed in 35 years.

One of the looming questions is: "What will happen when there is nothing that any human can do that isn't done thousands of times faster, cheaper, and better by machines?"

[Bob Kuczewski]

[Bob Kuczewski]

Moderator's Note:
   The videos originally included in this post appear
   to be no longer available. They were clips from the 
   "Twilight Zone" episode "Steel" which portrayed a 
   human boxer attempting to fight a robotic boxer.
   Rod Serling concluded the episode by stating:
   "Proof positive that you can't outpunch machinery."

Twilight Zone "Steel" Episode

Written by Richard Matheson (Based on his short story Steel, first published in the May 1956 edition of The Magazine of Fantasy and Science Fiction).

Original air date October 4, 1963.











"Proof positive that you can't outpunch machinery."
                                                      - Rod Serling

[Bob Kuczewski]

I have long feared that "self driving cars" will eventually lead to it becoming illegal for people to drive themselves. I figured it would happen gradually with insurance companies giving "incentives" to people using self-driving cars. Of course, an "incentive" to one group is another word for "penalty" to everyone else. Eventually it will become economically impractical for most people, and then it will become illegal. Doug's quote from Oz is right on point:

The end of human intelligence and coordination.
I loathe the day when the insurance companies will demand that every road vehicle shall be piloted by AI rather than a human. Performance cars and trucks will be a thing of the past, as will motorcycles, etc. If the gov't thinks you owe back taxes or doesn't like your politics, or some other made-up misdemeanor, guess where the vehicle will take you without your permission? End of freedom.

In my early career in AI, I didn't see any danger because I saw AI being used in isolated applications. I didn't foresee that so many aspects of our lives (banking, driving, voting, communication, ...) would be connected to a network of computers. Everything is in place now. Just a little more training and ...

[Bob Kuczewski]

https://www.thesocialdilemma.com/

The problem beneath all other problems

Technology’s promise to keep us connected has given rise to a host of unintended consequences that are catching up with us. If we can’t address our broken information ecosystem, we’ll never be able to address the challenges that plague humanity.

Linked article:

Opinion
TURNING POINTS

Our Brains Are No Match for Our Technology
By Tristan Harris
Dec. 5, 2019

Credit...Nicolas Ortega for The New York Times

Turning Point: In July, the Federal Trade Commission announced that it would fine Facebook $5 billion, the largest penalty ever levied by the agency for consumer privacy violations.

A decade ago, Edward O. Wilson, the Harvard professor and renowned father of sociobiology, was asked whether humans would be able to solve the crises that would confront them over the next 100 years.

“Yes, if we are honest and smart,” he replied. “The real problem of humanity is the following: We have Paleolithic emotions, medieval institutions and godlike technology.”

Since Mr. Wilson’s observation, technology’s godlike powers have increased dramatically, while the ancient, Paleolithic impulses of our brains have remained the same.

Yet this isn’t usually one of the complaints leveled against technology companies today — that the digital infrastructures of Facebook and Google have overwhelmed the natural capacities of our brains. Instead, we hear concerns that tech firms are collecting and tracking our personal data. Or that they’re simply too big.

Let’s imagine that we managed to solve the privacy issue. In this new utopia, we would own all our data, and tech giants would be forbidden from tracking our online whereabouts; they would have access only to the data we agreed to share.

While we might see fewer creepy ads and feel less paranoid about surveillance, the troubling trends connected to the online world would remain unaddressed.

Our addiction to social validation and bursts of “likes” would continue to destroy our attention spans. Our brains would still be drawn to outrage and angry tweets, replacing democratic debate with childlike he-said, she-said. Teenagers would remain vulnerable to online social pressure and cyberbullying, harming their mental health.

Content algorithms would continue to drive us down rabbit holes toward extremism and conspiracy theories, since automating recommendations is cheaper than paying human editors to decide what’s worth our time. And radical content, incubated in insular online communities, would continue to inspire mass shootings.

By influencing two billion brains in these ways, today’s social media holds the pen of world history: The forces it has unleashed will affect future elections and even our ability to tell fact from fiction, increasing the divisions within society.

Yes, online privacy is a real problem that needs to be addressed. But even the best privacy laws are only as effective as our Paleolithic emotions are resistant to the seductions of technology.

A viral app called FaceApp recently persuaded 150 million people to hand over private images of their faces, paired with their names, simply by appealing to their vanity. How? The app offered the ability to create surreally accurate portraits of people as they would look many years in the future. Who owns the app (and the 150 million names and faces)? A Russian company based in St. Petersburg.

Who needs to hack elections or steal voter information when people will happily hand over scans of their faces when you appeal to their vanity?

With our Paleolithic instincts, we’re simply unable to resist technology’s gifts. But this doesn’t just compromise our privacy. It also compromises our ability to take collective action.

That’s because our Paleolithic brains aren’t built for omniscient awareness of the world’s suffering. Our online news feeds aggregate all of the world’s pain and cruelty, dragging our brains into a kind of learned helplessness. Technology that provides us with near-complete knowledge without a commensurate level of agency isn’t humane.

Our Paleolithic brains also aren’t wired for truth-seeking. Information that confirms our beliefs makes us feel good; information that challenges our beliefs doesn’t. Tech giants that give us more of what we click on are intrinsically divisive. Decades after splitting the atom, technology has split society into different ideological universes.

Simply put, technology has outmatched our brains, diminishing our capacity to address the world’s most pressing challenges. The advertising business model built on exploiting this mismatch has created the attention economy. In return, we get the “free” downgrading of humanity.

This leaves us profoundly unsafe. With two billion humans trapped in these environments, the attention economy has turned us into a civilization maladapted for its own survival.

Here’s the good news: We are the only species self-aware enough to identify this mismatch between our brains and the technology we use. Which means we have the power to reverse these trends.

The question is whether we can rise to the challenge, whether we can look deep within ourselves and use that wisdom to create a new, radically more humane technology. “Know thyself,” the ancients exhorted. We must bring our godlike technology back into alignment with an honest understanding of our limits.

This may all sound pretty abstract, but there are concrete actions we can take.

First, policymakers can create a special tax for tech giants — a “downgrading tax” — that would make their business models, based on extracting and exhausting our attention spans, prohibitively expensive, while redistributing wealth to journalism, public education and the creation of new platforms that privilege human values and service to society.

Second, instead of joining free social media platforms that benefit from turning us into addicted, narcissistic extremists, we could agree to pay subscription fees to services that shun “likes” for features that empower our lives offscreen, making these services, in essence, fiduciaries acting in the best interests of humanity.

Third, instead of spreading disinformation, digital platforms could radically strengthen the media infrastructures that protect us from malicious viral content and tech-enabled distortions like “deepfakes” (fabricated videos manipulated by artificial intelligence to look genuine).

Candidates in the 2020 United States presidential election must educate themselves about the threat posed by technology’s race to outmatch our brains, and the news media must hold them accountable. No president can effectively deliver on his or her campaign promises without addressing the attention economy.

To create humane technology we need to think deeply about human nature, and that means more than just talking about privacy. This is a profound spiritual moment. We need to understand our natural strengths — our capacity for self-awareness and critical thinking, for reasoned debate and reflection — as well as our weaknesses and vulnerabilities, and the parts of ourselves that we’ve lost control over.

The only way to make peace with technology is to make peace with ourselves.

[Bob Kuczewski]

In my early career in AI, I didn't see any danger because I saw AI being used in isolated applications. I didn't foresee that so many aspects of our lives (banking, driving, voting, communication, ...) would be connected to a network of computers. Everything is in place now. Just a little more training and ...

One of the most dangerous aspects of artificial intelligence is its ability to communicate with other AI devices. Networking capability is being built into more and more of our consumer devices in the name of "convenience".

Frank shared an article with me about Amazon's "Sidewalk" from the Los Angeles Times. The article explains how many of Amazon's consumer products are programmed to network with other Amazon devices in your neighborhood.

https://enewspaper.latimes.com/infinity ... 8ce2fbf1ca

Echo and Ring may be sharing your Wi-Fi
The Amazon devices create networks in their neighborhoods.
Owners can opt out.

AMAZONLAUNCHED a program Tuesday that causes users of many Echo smart speakers and Ring security cameras to automatically share a small portion of their home wireless bandwidth with neighbors.

Amazon_Devices.png (172.83 KiB) Viewed 3764 times

By Joseph Pisani and Matt O’Brien

Do you own an Amazon smart device? If so, odds are good that the company is already sharing your internet connection with your neighbors unless you’ve specifically told it not to.

On Tuesday, the company launched a program that forces users of many Echo smart speakers and Ring security cameras to automatically share a small portion of their home wireless bandwidth with neighbors.

The way to stop it is to turn the option off.

Amazon says the program, called Amazon Sidewalk, is a way to make sure lights, smart locks and other gadgets outside the home and out of reach of a Wi-Fi connection stay working.

But some experts warn that the technology is so new that potential privacy and security risks remain unclear.

And almost no one seems happy that Amazon forced consumers into Amazon Sidewalk — or that many people may not know they can opt out of it.

Did Amazon notify customers?

Amazon says it sent emails to customers last month and in November that Sidewalk was coming. The company says you’ll also get a notification when you set up gadgets that work with Sidewalk.

How can I stop this?

Once you know about it, it’s relatively straightforward, if not exactly simple, to opt out of Sidewalk. Echo users can go into the Alexa app, tap “More” in the lower right corner, then tap “Settings,” then “Account Settings,” where they’ll find a section for Amazon Sidewalk and a button to disable it. In the Ring app, go to “Control Center” and then tap “Sidewalk.”

Why is Amazon doing this?

The idea behind Sidewalk is to integrate residential wireless connections into a “mesh network” that can extend coverage to areas that a home’s Wi-Fi can’t reach. Amazon’s Echo and Ring devices band together to create a network by grabbing a slice of bandwidth from each cooperative home network. That can extend the range of devices designed to work with Sidewalk so they’ll stay connected even when away from their owners’ home network.

One example is Tile, a tracking device that can be placed on keys or a dog’s collar. If your dog goes missing in a neighborhood where Sidewalk is working, it might turn up quickly via Tile.

Other products that work with Sidewalk include smart locks that can be controlled by phone and wearable devices that can track people with dementia. Amazon expects that additional devices, including outdoor lights and motion detectors, will work with Sidewalk before long.

How does it work?

Amazon is tapping into a variety of radio technologies, including one called LoRa for its long range and better known for industrial and business applications such as keeping track of cattle on pasturelands.

“The goal here is not to create coverage for a single home,” said Marc Pegulu of chipmaker Semtech, which is partnering with Amazon on the technology. “It’s kind of a shared network, a community shared network.”

Among other things, devices connected to the shared network can supposedly search for the best signal in a way that can preserve their battery life.

Thomas Clausen, a computer science professor at France’s École Polytechnique, said that the network Amazon is creating could be a good thing if it’s open and accessible because that would make it easier and cheaper to use internet-connected devices and sensors and spark new innovations.

“But of course, when a behemoth like Amazon, or Apple or Google or Microsoft, create a platform, they also can end up controlling the platform, and that can stifle innovation,” he said.

What are the risks?

Hackers could infiltrate home networks via Sidewalk, said Eric Null, a policy manager at digital rights group Access Now. If that happens, Null said, hackers could observe everything you do, take over your devices or access your files to steal information.

“It’s only a matter of time before someone’s network gets hacked and data gets breached,” Null said.

William Tong, Connecticut’s attorney general, warned this week that the program was “uncharted territory” and that people should opt out of Sidewalk unless they are sure their privacy and security will be protected. “Wireless networks are already notoriously vulnerable to hacks and breaches,” Tong wrote in a statement. “Families need better information and more time before giving away a portion of their bandwidth to this new system.”

What does Amazon say?

When asked about privacy and security concerns, an Amazon spokesperson said he couldn’t respond to “hypothetical situations.” The company didn’t respond to a request to make an executive available for an interview but said information that flows through its network has three layers of encryption that’s meant to keep hackers from seeing it.

Samir Jain, director of policy at the Center for Democracy and Technology, said Amazon’s effort to encrypt data was a good step. But Jain said it’s hard to identify security vulnerabilities until a new technology is deployed in the real world.

Which devices want to share my Wi-Fi?

Echo devices that share network bandwidth via Sidewalk include the third generation and newer versions of the Echo, Echo Dot, Echo Dot for Kids and Echo Dot With Clock. They also include the Echo Show 5, 8 and 10; the second generation of the Echo Show; Echo Spot; Echo Plus; Echo Studio; Echo Input; and Echo Flex.

Sidewalk-enabled Ring devices include the Ring Floodlight Cam, Ring Spotlight Cam Wired and Ring Spotlight Cam Mount.

Does Amazon have bigger ambitions?

Amazon has explained Sidewalk in the context of consumer devices such as cameras and speakers. But it could also one day integrate the technology into its core business: delivering goods.

For example, the company could embed tiny internet-connected devices into cardboard packages to help track them, said Clausen of École Polytechnique.

“Having integration of alarm systems, presence detection of whether or not people are there, will potentially allow Amazon to create more efficient delivery schedules and fewer re-deliveries,” he said. That could be one application Amazon is thinking about, he said.

[Bob Kuczewski]

This morning a friend sent me this article ("Hacking Humanity: Transhumanism" 04/14/2023 by Michael Rectenwald):

     https://mises.org/wire/hacking-humanity-transhumanism

Here are some quotes from that article:

Vernor Vinge, the mathematician, computer scientist, and science fiction author introduced the notion of the technological singularity in 1993. The singularity, Vinge suggested, is the near-future point at which machine intelligence will presumably supersede human intelligence. Vinge boldly declared: “Within thirty years, we will have the technological means to create superhuman intelligence. Shortly after, the human era will be ended.” Vinge predicted that the singularity would be reached no later than, you guessed it, 2030. The question Vinge addressed was whether, and if so, how, the human species might survive the coming singularity.

Harari peels back the curtain masking transhumanism’s Wizard of Oz promises, suggesting that even before the singularity, robotics and machine intelligence will make the masses into a new “useless class.” Given the exorbitant cost of entry, only the elite will be able to afford actual enhancements, making them a new, superior species—notwithstanding the claim that Moore’s Law closes the technological breach by exponentially increasing the price-performance of computing and thus halving its cost per unit of measurement every two years or less. How the elite will maintain exclusive control over enhancements and yet subject the masses to control technologies is never addressed. But perhaps a kill switch could be implemented such that the elite will not be subjected to brain-data mining—unless one runs afoul of the agenda, in which case brain-data mining could be (re)enabled.

In a 2018 WEF statement, Harari spoke as the self-proclaimed prophet of a new transhumanist age, saying:

We are probably among the last generations of homo sapiens. Within a century or two, Earth will be dominated by entities that are more different from us, than we are different from Neanderthals or from chimpanzees. Because in the coming generations, we will learn how to engineer bodies and brains and minds. These will be the main products of the 21st century economy.

No longer capable of mounting a challenge to the elite as in the nineteenth and twentieth centuries, and having no function, the feckless masses will have no recourse or purpose. Exploitation is one thing; irrelevance is quite another, says Harari. And thus, as Harari sees it, the remaining majority will be condemned to spend their time in the metaverse, or worse. If they are lucky, they will collect universal basic income (UBI) and will best occupy themselves by taking drugs and playing video games. Of course, Harari exempts himself from this fate.

As for the elite, according to Harari, their supposed superiority to the masses will soon become a matter of biotechnological fact, rather than merely an ideological pretension, as in the past. The elite will not only continue to control the lion’s share of the world’s material resources; they will also become godlike and enjoy effective remote control over their subordinates. Further, via biotechnological means, they will acquire eternal life on Earth, while the majority, formerly consoled by the fact that at least everybody dies, will now lose the great equalizer. As the supernatural is outmoded, or sacrificed on the altar of transhumanism, the majority will inevitably forfeit their belief in a spiritual afterlife. The theistic religions that originated in the Middle East will disappear, to be replaced by new cyber-based religions originating in Silicon Valley. Spirituality, that is, will be nothing but the expression of reverence for newly created silicon gods, whether they be game characters, game designers, or the elites themselves.

Harari’s pronouncements may amount to intentional hyperbole to make a point, but his statements are remarkable for the cynicism and disdain for humanity they betray. They are revelatory of the unmitigated gall of believers in the transhuman future. Coupled with the neo-Malthusian impulses of the elite, centered around the UN and the WEF, a picture emerges of an elite whose objective is to reduce the population of “useless eaters,” while keeping the remainder in their thrall.

[This piece is an excerpt from The Great Reset and the Struggle for Liberty.]

[Bob Kuczewski]

https://www.humanetech.com/

[Bob Kuczewski]

https://www.washingtonpost.com/business ... story.html

Carmakers fail privacy test, give owners little or no control on personal data they collect
By Frank Bajak | AP
September 6, 2023 at 2:00 a.m. EDT

BOSTON — Cars are getting an “F” in data privacy. Most major manufacturers admit they may be selling your personal information, a new study finds, with half also saying they would share it with the government or law enforcement without a court order.

The proliferation of sensors in automobiles — from telematics to fully digitized control consoles — has made them prodigious data-collection hubs.
But drivers are given little or no control over the personal data their vehicles collect, researchers for the nonprofit Mozilla Foundation said Wednesday in their latest “Privacy Not Included” survey Security standards are also vague, a big concern given automakers’ track record of susceptibility to hacking.

“Cars seem to have really flown under the privacy radar and I’m really hoping that we can help remedy that because they are truly awful,” said Jen Caltrider, the study’s research lead. “Cars have microphones and people have all kinds of sensitive conversations in them. Cars have cameras that face inward and outward.”
Unless they opt for a used, pre-digital model, car buyers “just don’t have a lot of options,” Caltrider said.
Cars scored worst for privacy among more than a dozen product categories — including fitness trackers, reproductive-health apps, smart speakers and other connected home appliances — that Mozilla has studied since 2017.
Not one of the 25 car brands whose privacy notices were reviewed — chosen for their popularity in Europe and North America — met the minimum privacy standards of Mozilla, which promotes open-source, public interest technologies and maintains the Firefox browser. By contrast, 37% of the mental health apps the non-profit reviewed this year did.

Nineteen automakers say they can sell your personal data, their notices reveal. Half will share your information with government or law enforcement in response to a “request” — as opposed to requiring a court order. Only two — Renault and Dacia, which are not sold in North America — offer drivers the option to have their data deleted.
“Increasingly, most cars are wiretaps on wheels,” said Albert Fox Cahn, a technology and human rights fellow at Harvard’s Carr Center for Human Rights Policy. “The electronics that drivers pay more and more money to install are collecting more and more data on them and their passengers.”
“There is something uniquely invasive about transforming the privacy of one’s car into a corporate surveillance space,” he added.

A trade group representing the makers of most cars and light trucks sold in the U.S., the Alliance for Automotive Innovation, took issue with that characterization. In a letter sent Tuesday to U.S. House and Senate leadership, it said it shares “the goal of protecting the privacy of consumers.”

It called for a federal privacy law, saying a “patchwork of state privacy laws creates confusion among consumers about their privacy rights and makes compliance unnecessarily difficult.” The absence of such a law lets connected devices and smartphones amass data for tailored ad targeting and other marketing — while also raising the odds of massive information theft through cybersecurity breaches.
The Associated Press asked the Alliance, which has resisted efforts to provide car owners and independent repair shops with access to onboard data, if it supports allowing car buyers to automatically opt out of data collection — and granting them the option of having collected data deleted. Spokesman Brian Weiss said that for safety reasons the group “has concerns” about letting customers completely opt out — but does endorse giving them greater control over how the data is used in marketing and by third parties.

In a 2020 Pew Research survey, 52% of Americans said they had opted against using a product or service because they were worried about the amount of personal information it would collect about them.
On security, Mozilla’s minimum standards include encrypting all personal information on a car. The researchers said most car brands ignored their emailed questions on the matter, those that did offering partial, unsatisfactory responses.
Japan-based Nissan astounded researchers with the level of honesty and detailed breakdowns of data collection its privacy notice provides, a stark contrast with Big Tech companies such as Facebook or Google. “Sensitive personal information” collected includes driver’s license numbers, immigration status, race, sexual orientation and health diagnoses.

Further, Nissan says it can share “inferences” drawn from the data to create profiles “reflecting the consumer’s preferences, characteristics, psychological trends, predispositions, behavior, attitudes, intelligence, abilities, and aptitudes.”
It was among six car companies that said they could collect “genetic information” or “genetic characteristics,” the researchers found.
Nissan also said it collected information on “sexual activity.” It didn’t explain how.
The all-electric Tesla brand scored high on Mozilla’s “creepiness” index. If an owner opts out of data collection, Tesla’s privacy notice says the company may not be able to notify drivers “in real time” of issues that could result in “reduced functionality, serious damage, or inoperability.”

Neither Nissan nor Tesla immediately responded to questions about their practices.
Mozilla’s Caltrider credited laws like the 27-nation European Union’s General Data Protection Regulation and California’s Consumer Privacy Act for compelling carmakers to provide existing data collection information.
It’s a start, she said, by raising awareness among consumers just as occurred in the 2010s when a consumer backlash prompted TV makers to offer more alternatives to surveillance-heavy connected displays.

[Bob Kuczewski]

This video is titled "How Microsoft Took Control of OpenAI".

The one sentence answer is "Follow the money".

https://www.youtube.com/watch?v=o07kRSvz7yM