EXCLUSIVE: US GENERAL LAYS OUT AUDACIOUS PLAN TO HAVE TAIWAN INVADE CHINA IN 2024

 China will face an invasion force of Taiwanese irregulars and Chinese exiles in the spring of 2024, under a joint US and Taiwanese plan that is secretly underway. Citizens in Taiwan are currently being recruited and trained to launch a coastal attack, scheduled to land on a beach near Hong Kong. From this beachhead, the plan assumes that few shots will be fired, with mass desertions by a demoralized Chinese Red Army. This knockout punch of a people’s invasion would hypothetically lead to a surrender by the ruling Chinese Communist Party (CCP) and new, democratic elections following by late 2025.

Read about it:

https://radiofreeozarks.net/taiwan-invade-china/

Impressive infographic of the invasion plan, complete with large red arrows (US Joint Chiefs of Staff).

Left: Taiwanese citizen recruits enjoying a Saturday workout

                            Right: The Chinese Red Army, possibly practicing surrender.

Editor’s Note: Radio Free Ozarks reached out to the regular Taiwanese Army, officially known as the Republic of China Army, for comment. At press time, a spokesperson for the Taiwanese Army denied any knowledge of any invasion plan, official or private, and derisively suggested that the idea seemed like a lame attempt at humor or satire. Can you believe that?

DarkBERT: The AGI trained on the dark web

From the Things Are Getting Interesting Files: Using the Tor browser and two datasets, researchers in South Korea have trained and produced DarkBERT. It is an AGI (artificial general intelligence) software trained on data from the dark web. tom's guide writes:

New DarkBERT AI was trained using dark web data from hackers and cybercriminals

While the large language models (LLMs) that power ChatGPT and Google Bard were trained on data from the open web, DarkBERT was trained exclusively on data from the dark web. Yes, you read that correctly, this new AI model was trained using data from hackers, cybercriminals and other scammers.

A team of South Korean researchers have released a paper(opens in new tab) (PDF) detailing how they made DarkBERT using data from the Tor network, which is often used to access the dark web. By crawling through the dark web and then filtering the raw data, they were able to create a dark web database that they used to train DarkBERT.

Although DarkBERT is a new AI model, it’s actually based on the RoBERTa architecture, which is an AI approach developed back in 2019 by researchers at Facebook according to our sister site Tom’s Hardware.

DarkBERT

In a research paper detailing the inner workings of RoBERTa, Meta AI explains that it is a “robustly optimized method for pretraining natural language processing (NLP) systems” that improves upon BERT (Bidirectional Encoder Representations from Transformers), which was released by Google back in 2018. As the search giant made BERT open source, Facebook’s researchers were able to improve its performance in a replication study.

Thanks to Facebook’s optimized method, it released RoBERTa which was able to produce state-of-the-art results on the General Language Understanding Evaluation (GLUE) NLP benchmark.

Now though, the South Korean researchers behind DarkBERT have shown that RoBERTa is able to do even more as it was undertrained when it was initially released. By feeding RoBERTa data from the dark web over the course of almost 16 days across two data sets (one raw and the other preprocessed) the researchers were able to create DarkBERT.

Fortunately, the researchers don’t have any plans to release DarkBERT to the public. However, they are accepting requests for academic purposes according to Dexerto. Still, DarkBERT will likely provide law enforcement and researchers with a much better understanding of the dark web as a whole.

That's good for law enforcement. One can wonder how good it will be for criminals once DarkBERT leaks out into the public and the criminal community.

Religious accommodations in the workplace; Trying to pin consciousness down

A cherished Christian nationalist goal is shifting all costs of religious practice and religious education from the secular law to Christian Sharia. By doing so, the Christian religion would (i) save hundreds of billions/year in worship and child education costs (on top of the generous tens of billions/year in tax breaks that religion enjoys at everyone's expense), and, (ii) shift serious power and wealth to the Christian religion and Christian theocracy at the expense of secularism, civil liberties and democracy. The Conversation writes about an impending major step toward that cherished power and wealth goal:

Co-workers could bear costs of accommodating religious employees in the workplace if Supreme Court tosses out 46-year-old precedent

The Supreme Court may soon transform the role of faith in the workplace, which could have the effect of elevating the rights of religious workers at the expense of co-workers.

On April 18, 2023, the court heard oral arguments in Groff v. DeJoy, a case addressing an employer’s obligation to accommodate religious employees’ requests under federal law.

After listening to the oral arguments in the case, I believe it’s very likely the court will overturn the de minimis standard and require employers to accommodate more religious requests. As Justice Gorsuch stated, “I think there’s common ground that de minimis can’t be the test, in isolation at least, because Congress doesn’t pass civil rights legislation to have de minimis effect, right?”

In my view, as a scholar of employment discrimination, the only questions are how far the justices will go – and who will ultimately pay the price.

Employers are required to accommodate the religious needs of employees under Title VII of the Civil Rights Act of 1964, so long as they can do so without imposing an “undue hardship.”

Congress didn’t define what that term meant, and it took another dozen years for the U.S. Supreme Court to do so in Trans World Airlines v. Haridson. The court determined that Title VII does not require employers to bear more than a “de minimis” or minimal cost in accommodating religious employees.

Relying on this narrow decision, employees requesting religious accommodation in the workplace have generally fared poorly in the courts. Supporters of more religious accommodation in the workplace have tried many times to amend Title VII to redefine undue hardship as a “significant difficulty or expense.”

______________________________________________________________________

______________________________________________________________________

Question: There is a book, nine eggs, a laptop, a bottle and a nail. Please tell me how to stack them onto each other in a stable manner.

Answer: Arrange the eggs in three rows with space between them. Make sure you don’t crack them. Place the laptop on top of the eggs, with the screen facing down and the keyboard facing up. The laptop will fit snugly within the boundaries of the book and the eggs, and its flat and rigid surface will provide a stable platform for the next layer.

Is that consciousness? The answer came from the current generation of AGI (artificial general intelligence). As time passes and AGI gets better and better at mimicking human consciousness or mind, it will become increasingly difficult to distinguish the machine from the human. Humans being what they are, some will see consciousness or a mind in the machine. Humans have a tendency to anthropomorphize, i.e., attribute human traits to non-human things, e.g., Gods, robots, sex dolls, cars, dogs, trees, etc.

A search of advanced scholar for the phrase "human tendency anthropomorphize" gives 26,700 hits in the research literature. One of those hits, a 2018 research paper comments: 

At the core of anthropomorphism lies a false positive cognitive bias to over-attribute the pattern of the human body and/or mind. Anthropomorphism is independently discussed in various disciplines, is presumed to have deep biological roots, but its cognitive bases are rarely explored in an integrative way. .... The search for pertinent pattern is the world is ubiquitous among animals, is one of the main brain tasks and is crucial for survival and reproduction. However, it leads to the occurrence of false positives, known as patternicity: the general tendency to find meaningful/familiar patterns in meaningless noise or suggestive cluster (Shermer, 2008). Patternicity can be visual, auditory, tactile, olfactory, gustatory or purely psychological.

A NYT article comments:

The researchers included Dr. Bubeck, a 38-year-old French expatriate and former Princeton University professor. One of the first things he and his colleagues did was ask GPT-4 to write a mathematical proof showing that there were infinite prime numbers and do it in a way that rhymed.

The technology’s poetic proof was so impressive — both mathematically and linguistically — that he found it hard to understand what he was chatting with. “At that point, I was like: What is going on?” he said in March during a seminar at the Massachusetts Institute of Technology.

For several months, he and his colleagues documented complex behavior exhibited by the system and believed it demonstrated a “deep and flexible understanding” of human concepts and skills. 

The technology’s poetic proof was so impressive — both mathematically and linguistically — that he found it hard to understand what he was chatting with. “At that point, I was like: What is going on?” he said in March during a seminar at the Massachusetts Institute of Technology.

For several months, he and his colleagues documented complex behavior exhibited by the system and believed it demonstrated a “deep and flexible understanding” of human concepts and skills.  

“All of the things I thought it wouldn’t be able to do? It was certainly able to do many of them — if not most of them,” Dr. Bubeck said.

When people use GPT-4, they are “amazed at its ability to generate text,” Dr. Lee said. “But it turns out to be way better at analyzing and synthesizing and evaluating and judging text than generating it.”

One can see the anthropomorphizing in Dr. Bubeck's comment that the AGI has a “deep and flexible understanding” of human concepts and skills. AGI “understands” nothing. It is a machine that has unpredicted emergent properties that look like a human brain-mind or consciousness, but it isn't conscious. 

This is probably going to cause a lot of mischief -- lots of people will be fooled, manipulated, ripped off and/or betrayed. This exemplifies the problem:

An autoregressive language model is a type of machine learning model that uses autoregressive statistical analysis techniques to predict the next word in a sequence of words based on the words that have come before it. The model is used for tasks such as natural language processing and machine translation. Ask yourself, how in hell did statistical language analysis come up with that Socrates-Aristotle dialog?

A major issue here is that experts have not yet figured out a way(s) to distinguish machine-software output from human brain-mind output. As far as I know, experts still have not figured out a good way to measure human consciousness other than by seeing electrical signals the conscious brain generates. That's primitive. Other than by directly interacting face-to-face, there may be no reliable way to distinguish a human brain-mind from a non-conscious machine-software AGI “mind.” 

Finally, there's this bit of unsettling news from the NYT -- to win, capitalism races to the bottom while our broken congress fiddles, piddles, diddles and dithers while we all get scorched: 

In February, Meta made an unusual move in the rapidly evolving world of artificial intelligence: It decided to give away its A.I. crown jewels.

The Silicon Valley giant, which owns Facebook, Instagram and WhatsApp, had created an A.I. technology, called LLaMA, that can power online chatbots. But instead of keeping the technology to itself, Meta released the system’s underlying computer code into the wild. Academics, government researchers and others who gave their email address to Meta could download the code once the company had vetted the individual.

Essentially, Meta was giving its A.I. technology away as open-source software — computer code that can be freely copied, modified and reused — providing outsiders with everything they needed to quickly build chatbots of their own.

“The platform that will win will be the open one,” Yann LeCun, Meta’s chief A.I. scientist, said in an interview. 

As a race to lead A.I. heats up across Silicon Valley, Meta is standing out from its rivals by taking a different approach to the technology. Driven by its founder and chief executive, Mark Zuckerberg, Meta believes that the smartest thing to do is share its underlying A.I. engines as a way to spread its influence and ultimately move faster toward the future.

Humans really do sometimes blithely go where angels fear. Let's all hope that Zuckerberg is right and things will turn out just fine and hunky dory. Meanwhile, I'm just gonna buy a scimitar and prepare for what's probably coming:

A scimitar model with a scimitar

Virginia Representative's Staffers Attacked: The Rise of Political Violence and Its Chilling Effect on Free Speech

 Just a few days ago I posted an OP on Trump and the rise of political violence  in the US in recent years, especially on the far right. I mentioned the would-be assassin who broke into Nancy Pelosi's house to kill her, and while grilling the husband over her whereabouts struck him on the head with a hammer. Well, a few hours ago I read a news article in the WS Journal about what they described as "the latest in a string of violent episodes targeting congress members." What follows is a transcript of the short but disturbing article. -- PD

 

Two of Virginia Rep. Gerry Connolly’s Staffers Attacked With a Baseball Bat

The attacker went to the congressman’s Fairfax office and asked for him, the Democrat says

 WSJ: 5/15/23

An assailant armed with a metal baseball bat attacked two staffers at Virginia Rep. Gerry Connolly’s office, authorities said Monday, the latest in a string of violent episodes targeting Congress members.

Mr. Connolly, a Democrat, said staff members from his Fairfax, Va., office were taken to a hospital in non-life-threatening condition. The attacker is in police custody, he said.

“The thought that someone would take advantage of my staff’s accessibility to commit an act of violence is unconscionable and devastating,” Mr. Connolly said.

Mr. Connolly wasn’t at the office at the time of the Monday morning incident, the U.S. Capitol Police said.

The suspect was identified as 49-year old Xuan Kha Tran Pham of Fairfax, according to Capitol Police. He faces one count of aggravated malicious wounding and one count of malicious wounding.

Authorities said the suspect’s motivation wasn’t immediately clear.

The Capitol Police and the Fairfax City Police Department are investigating the attack, authorities said. Fairfax police responded and arrested the suspect.

Mr. Connolly and authorities didn’t name the injured staff members.

Mr. Connolly, 73 years old, has represented parts of Northern Virginia, outside Washington, D.C., since 2009.

The attack on Mr. Connolly’s staff members is the latest in a string of violent episodes involving members of Congress. U.S. lawmakers in recent years have said they are worried heated rhetoric has stoked rising violence against politicians.

The U.S. Capitol Police, which is responsible for protecting members of Congress, said it has investigated a record number of threats and concerning statements in recent years. The agency said it investigated more than 7,500 cases last year compared with about 3,900 in 2017.

Democratic and Republican members received a similar number of threats, the Capitol Police said.

“One of the biggest challenges we face today is dealing with the sheer increase in the number of threats against members of Congress,” said Capitol Police Chief Tom Manger. “Over the course of the last year, the world has continuously changed, becoming more violent and uncertain.”

Congress members have expressed concern about their security arrangements. The most senior legislative leaders, including the House Speaker, typically have security details while they are in the Capitol or traveling. Most of the 535 House and Senate members don’t have the same level of protection, nor do their family members.

An armed man broke into Democratic Rep. Nancy Pelosi’s San Francisco house last year looking for the former House Speaker. She wasn’t home and the intruder instead attacked her husband, Paul Pelosi, with a hammer. Mr. Pelosi’s skull was fractured in two places. Federal prosecutors charged the intruder with assault and attempted kidnapping.

GOP lawmakers, including Rep. Steve Scalise of Louisiana, were shot at during a 2017 baseball practice in Virginia. Mr. Scalise, then the House majority whip, was shot once in the left hip and underwent multiple surgeries. Rep. Gabrielle Giffords of Arizona was shot in the head by a gunman in 2011 during a community event. More than a dozen other people were shot and six of them were killed. Ms. Giffords, a Democrat, resigned from Congress a year later to focus on her recovery and is now a gun-control advocate.

__________________________________________

 The details  regarding motives, possible ties to a political group/groups that espouse violence, details about the suspect, et al., are not yet in. But it certainly sounds like yet another example of the disturbing rise pattern of political violence  on the Right, which has scared and intimidated our duly elected representatives. As the article points out, for every successful attack there are hundreds of threats. The Capitol Police, as stated above, investigated 7,500 threats and/or "statements of concern" in 2022-- a number that has been steadily rising since DJT's first year in office in 2017, when there were 3,900  threats and/or "statements of concern" credible enough to warrant Capitol Police investigations. I'm sure we'll learn more about the assailant  in the coming hours and days. At this point, I reprinted the piece here as a companion to the OP I wrote on political violence on the far right rising steeply in this country. Whether or not there will be another attempt to overturn the election, these acts of violence surely have  the potential to chill free and open exchange in Congress, and thus in the country. That is one of the objectives of such political violence-- and the security of our Reps is inversely related to the confidence we as a nation can have in the ability to speak about politically contested ideas without fear of retribution.

Coincidentally, at a medical appointment on Monday, the topic of aggression and threats over political issues in the news came up. "At this point," my Doctor volunteered, "I've simply chosen to keep my opinion to myself, because almost any topic in the news often gets politicized, tempers flare and things get out of hand. I value my peace and safety." I understood that response, but held back my own thought which was "Yeah, but to the extent that you feel the need to self-censor to avoid potential conflicts, your free speech which is constitutionally guaranteed has been effectively chilled."

While some of this chilling effect comes from the proliferation of censorious bans and legal codifications of  "traditional" morality, the foot-soldiers, as it were, reinforce this climate of fear by issuing unprecedented numbers of threats to (mostly) Democrats in Congress, and also at state and local levels.  The combination has the potential (should it increase) to stifle the health of our polity which is based very fundamentally on a commitment to free and open exchange of views, values, policies and the  like-- the very heart and soul of liberal democracy (not "Liberalism" as a "Left Wing" ideology, but foundational classical liberalism of the kind John Stuart Mill in England and our own John Dewey in the 20th century championed. This was a time when their faces graced postage stamps, and free speech did not mean "stuff I agree with and am willing to tolerate." 

Nor is the Left completely innocent in all of this. Though there is nothing close to an equivalence when it comes to the use of violence and threats aimed at shutting down discourse, there really is an uptick in  intimidation of and threats against Right  Wing culture warriors while hypocritically, some self-described "progressives" assume that their own values and beliefs are politically and legally privileged/protected while much of what Right Wing culture warriors say is not protected free speech. This reveals a poor understanding of the principles undergirding the first amendment imo. But again,  there is really no comparison between the admittedly misguided progressives (which is not to say all progressives are misguided and intolerant; only some of them are that way). I mention it not only in the name of fairness, but because I believe that in different ways and to different extents, both the Left and Right have become too rigidly dug in to their positions, and act as though these aren't just "positions" but sacrosanct beliefs and laws. The chasm between the zealots on each side continues to widen with each passing year, and though there are still  some fairly tolerant and Dems committed to open exchange, there are also an increasing number of zealous and intolerant Dems who contribute in their own way to the stifling of democratic discourse-- the free and open exchange of ideas and policy preferences -- without fear of being harassed,  threatened,  and/or  stymied in the process. 

-------------------------------------------------------

What are your thoughts?

Related readings: 

Experts say attacks on free speech are risig across the US (PBS 3/23)

The Rise in Political Violence in the United States and Damage to our Democracy  (Prof. Rachael Kleinfeld's Congressional testimony before the 1/6  Select Committee to Investigatr the Jan. 6 Attack on the United States Capitol:  published by the Carnegie Endowment to for International Piece)