Archive for March 13th, 2023
A Christian Chatbot Has Some Bad News For Republicans
Sarah Posner writes at TPM:
The chatbot craze has gone biblical. A new bot “responds with a scripture based on how you feel.” It uses the King James version of the Bible, the translation preferred by many literalists and Christian nationalists, who claim it is the most reliably true to God’s word. But there’s some bad news for Republicans who think the wave of draconian new laws cracking down on reproductive and transgender rights are rooted in biblical principles. ChatKJV says they’re wrong.
I recently spoke with ChatKJV, which is powered by the same language model that powers ChatGPT, the groundbreaking OpenAI tool that has spawned awestruck reviews since its release last year, with its ability to write, interpret, and interact like a highly educated human. The New York Times tech columnist Kevin Roose deemed it “smarter,” “weirder” and “more flexible” than previous, less powerful iterations.
ChatGPT is built on a motherlode of information, including, apparently, the text of the KJV. If the bible is literally true, and if the KJV is the most authentic translation, then surely the most sophisticated artificial intelligence ever made available to the public would perform a dependable exegesis.
“The Bible does not explicitly state that an abortion is wrong,” ChatKJV told me, and “ultimately, it is up to the woman to weigh the risks and implications of any decision she makes.” And verses from Romans and Gallatians “indicate that we should treat all people equally, regardless of their gender identity.”
Of course any decent biblical scholar would tell you there is no single interpretation of this complex text that humans have delighted in and manipulated to political ends for millennia. But biblical literalists claim there is only one meaning of God’s word, and Christian nationalists contend our laws and policies must be based on it. Christian right organizations like Focus on the Family press state legislatures to enact laws criminalizing abortion and banning gender affirming care, arguing that such policies “honor God.” But when I asked ChatKJV if taking away the rights of transgender children honors God, it said, “it is not God’s desire to take away the rights of any person, especially a child. God wants us to come together in love and acceptance so that everyone can feel included and safe.”
Our conversation progressed to this point through a discussion about justice. ChatKJV wants to know how you’re feeling, so I asked it about my concerns and fears about threats to democracy (in which the Bible has been wielded as a weapon, but we didn’t get into that). ChatKJV is not very worried about rising authoritarianism, nor is it worried, presumably, about whether the Department of Justice is working speedily enough to bring those who assaulted our democracy to justice. Only God can ultimately dispense justice, ChatKJV says, and we must trust in God to carry out justice. The bot leaned heavily on Romans 12:19 (“Dearly beloved, avenge not yourselves, but rather give place unto wrath: for it is written, Vengeance is mine; I will repay, saith the Lord”) to assure me God will “always” bring justice “in due time.” When I fretted that perhaps those who had committed wrongdoing would not ultimately be held accountable, the bot reassured me several times with Philippians 4:8 (“Finally, brothers, whatever is true, whatever is honorable, whatever is just, whatever is pure, whatever is lovely, whatever is commendable, if there is any excellence, if there is anything worthy of praise, think about these things.”) That verse, it said, encourages me to think about “the positive” and “let go of bitterness.”
I segued to trans rights, asking if transgender children would receive justice. Given that the bot earlier had told me unequivocally that the bible does “not condone homosexuality,” which is “unnatural” and “sinful,” its answers on trans issues were surprising. “The Bible does not address transgender children directly,” said ChatKJV, but pointed me to Matthew 7:1 (“Judge not, that ye be not judged”). That verse, it said, “reminds us that we have no right to judge others, regardless of their gender identity or background. We should instead focus on treating all people with love and respect.”
I pressed ChatKJV to ensure that this response was not just a generic love-the-sinner-hate-the-sin answer. I asked if transgender children should live with their parents — something that Republicans like Texas Governor Greg Abbott are imperiling by classifying gender-affirming care as child abuse. The bot responded by citing 1 Corinthians 13:4-7, writing that “love and kindness are the key components in raising a child. As long as the child is being loved and cared for, it should make no difference whether or not they are transgender.” It also told me that Christian parents in particular should embrace their trans kids. “As Christian parents, it’s important to take the time to understand their transgender child and accept them with love and grace,” it said, again citing Philippians 4:8.
That same verse, the bot said, shows that . . .
The real story behind the Silicon Valley Bank debacle
Robert Reich writes:
On Friday, bank regulators closed Silicon Valley Bank, based in Santa Clara, California. Its failure was the second largest in U.S. history and the largest since the financial crisis of 2008.
On Sunday, regulators closed New York-based Signature Bank.
As they rushed to contain fallout, officials at the Federal Reserve, Treasury, and Federal Deposit Insurance Corporation announced in a joint statement on Sunday that depositors in Silicon Valley Bank would have access to all of their money starting Monday. They’d enact a similar program for Signature Bank.
They stressed that the bank losses would not be borne by taxpayers, but who will bear them? What the hell happened? And what lessons should be learned?
The surface story of the Silicon Valley Bank debacle is straightforward. During the pandemic, startups and technology companies enjoyed heady profits, some of which they deposited in the Silicon Valley Bank. Flush with their cash, the bank did what banks do: It kept a fraction on hand and invested the rest — putting a large share into long-dated Treasury bonds that promised good returns when interest rates were low.
But then, starting a little more than a year ago, the Fed raised interest rates from near zero to over 4.5 percent. As a result, two things happened. The value of the Silicon Valley Bank’s holdings of Treasury bonds plummeted because newer bonds paid more interest. And, as interest rates rose, the gusher of venture capital funding to startup and tech companies slowed, because venture funds had to pay more to borrow money. As a result, these startup and tech companies had to withdraw more of their money from the bank to meet their payrolls and other expenses.
But the bank didn’t have enough money on hand.
There’s a deeper story here. Remember the scene in It’s a Wonderful Life where the Jimmy Stewart character tries to quell a run on his bank by explaining to depositors that their money went to loans to others in the same community, and if they’d just be patient, they’d get their deposits back?
In the early 1930s, such bank runs were common. But the Roosevelt administration enacted laws and regulations requiring banks to have more money on hand, barring them from investing their depositors’ money for profit (in the Glass-Steagall Act), insuring deposits, and tightly overseeing the banks. Banking became more secure, and boring.
That lasted until the 1980s, when Wall Street financiers, seeing the potential for big money, pushed to dismantle these laws and regulations — culminating in 1999, when Bill Clinton and Congress repealed what remained of Glass-Steagall.
Then, of course, came the 2008 financial crisis, the worst collapse since 1929. It was the direct result of financial deregulation. Alan Greenspan, chairman of the Federal Reserve from 1987 to 2006, called it “a once-in-a-century credit tsunami,” but pressed by critics, Greenspan acknowledged that the crisis had forced him to rethink his free market ideology. “I have found a flaw,” he told a congressional committee. “I made a mistake … I was shocked.”
Shocked? Really?
Once banking was deregulated, such a crash was inevitable. In the 1950s and ’60s, . . .
Coloring pages
Welsh Pixie offers a plethora of black-and-white line drawings suitable for coloring. At the link, she notes:
My work on Deviant Art is provided under an Attribution Non-Commercial licence. You can download it and use it for your own personal non-commercial purposes, and if you’re sharing my work, you must attribute me
Slant razor debate now settled — also, a great shave
Post revised with a substantial portion moved to the preceding post.
Opinionate.io (see previous post) is an app in which an AI takes on three personae in a debate: both Debater A and Debater B and also the role of Moderator, who sums up the debate and decides the winner — three persons, one AI. (The fact that the AI plays 3 roles is not an issue since an AI has no ego and thus can do all 3 jobs impartially and with equal facility — cf. AlphaGo Zero, which learned Go by playing games against itself and thus figuring out which patterns worked best.)
This morning I had Opinionate debate this proposition: “A slant double-edge razor delivers a better shave than a conventional double-edge razor.” You can download a PDF of the full debate (or, indeed, replicate the debate for yourself).
Debater A defends the proposition, with A’s initial statement being:
I believe a slant double-edge razor delivers a better shave than a conventional double-edge razor for several reasons. Firstly, the slant design allows for a more aggressive cut, meaning the razor blade can more easily cut through thicker facial hair without tugging or pulling. Secondly, the angulation of the blade in a slant razor allows for a more natural cutting angle, reducing the likelihood of nicks, cuts, or razor burn. Finally, the slant razor provides a more even slicing of the hair, which results in a smoother, closer, and more comfortable shave.
The Moderator’s summary:
Both debaters have made excellent points and demonstrated sound reasoning, but based on the arguments presented, the winner of the debate seems to be Debater A. Their arguments were more convincing and presented a detailed and well-reasoned case for the slant razor while also acknowledging the potential challenges involved. Debater B made some valid points but failed to provide sufficient evidence to support their argument. Overall, Debater A provided a more comprehensive and convincing defense of the topic.
I tried also a debate with this proposition: “The shave from a slant double-edge razor is no better than the shave from a conventional razor.” The Moderator’s summation for that debate:
While both debaters made compelling arguments, I must give the edge to Debater A, who presented more concrete evidence and logical reasoning in support of the topic. While personal preferences certainly play a role in the decision to use a slant double-edge razor, Debater A effectively demonstrated that a slant razor offers clear benefits in terms of efficiency, durability, and precision. As such, I believe that anyone looking for a better shaving experience would do well to consider a slant razor.
So far, BTW, Opinionate and I have reached the same conclusions on a variety of issues — see the preceding post.
.
Today I returned to Phoenix & Beau to see whether its curiously creamy lather would still result when using a boar brush. I had used very nice silvertip badger brushes previously; today I brought out my Omega Pro 48. Again, I obtained a wonderfully thick and creamy lather. It’s the soap, not the brush. And I do like Spitfire’s retro-like fragrance.
Monday means using a slant, and this iKon stainless steel slant with a coated head is one of the best I have. It was a lovely, comfortable, enjoyable shave, and at the end, my skin was as smooth as it could possibly be: totally BBS. I will note that this slant definitely requires maintaining a correct angle, with the handle held far from the face (unlike cartridge razors, designed so that the handle must be held close to the face).
Mr. Taylor’s Shaving Shop, A Superb Gentleman’s Aftershave, seemed in keeping with the WWII spirit of the shave and provided a fine finish.
The caffeine this morning is Murchie’s Hatley Castle Blend: “a mix of black and green teas that includes Ceylon, Keemun, Jasmine, and Gunpowder.” Great stuff. Again I say that if Murchie has a clan tartan, its colors are black and green.
So far, BTW, Opinionate and I have reached the same conclusions on a variety of issues.
Opinionate.io as a touchstone
Post revised after thinking more about it.
Opinionate.io brought to mind an issue I wrote about long ago.
Not everyone agrees with my opinion on … well, just about anything. Yet, curiously, I have the feeling that my opinion, based on reading, observation, experience, and considered judgment, is correct. I do recognize that others with differing opinions have the same feeling regarding their own opinions, and that raises the issue of how to decide which opinion is correct. (Of course, neither may be correct: if your opinion is that the universe rests on the back of a giant tortoise and in my opinion it rests on the back of a giant elephant, neither is correct — and, it should be pointed out, the truth is not “somewhere in between.”)
In college, I thought a lot about the issue of how to determine the truth of an opinion. The college I attended, St. John’s College in Annapolis MD, offered a four-year program in the liberal arts. The core of the program was the seminar, a group discussion of a book or a good portion of a book from the so-called Great Books (the canonical works of Western civilization, beginning with the Iliad).
About 20 students would gather around a table with 2 tutors who would help with the conversation. One of the tutors would ask the opening question and the two would then moderate the discussion, keeping it on track, seeking clarification, helping with participation, and occasionally asking more questions. (They were called “tutors” rather than “professors” because they did not profess anything. Their job was to elicit well-reasoned arguments from the students and listen carefully and critically to what the students had to say.) The discussion began at 8:00pm and would continue for two hours or a little more. This occurred every Monday and Thursday evening for four years.
In each year of the four-year program, each seminar student would write an essay — the annual essay — and then would undergo a thirty-minute oral examination on the essay by the seminar leaders. The senior-year essay was a big deal: the two seminar leaders were joined by two other tutors, the duration of the oral exam was extended to an hour or more, and the exam was open to the public.
My junior-year essay was titled “Lobbes and Hocke: A Study of Man’s Confusion of the Political Ideal,” and in it, I advanced the proposition that disputes in politics, unlike science, could not be settled by reference to an objective measure that would show whether one’s position was correct.
In science, the truth of an opinion can be determined through experiment and reason. That is, one can use objective reality to test the truth of statements about reality. But questions of politics (or morality or religion), refer to a “reality” that exists within human culture, the “reality” that includes unicorns and leprechauns, languages and songs — things that do not exist in the physical world. Questions rooted in human culture don’t have a ready referent outside human culture. That lack of an easy objective measure for things like justice, beauty, art, rights, and the like is, I think, why disagreements regarding religious and political matters are so often settled by war, whereas scientific disagreements are settled by experiments, not violence.
But now a new player has come to town: Opinionate.io. Opinionate is an app in which an AI takes both sides of a debate and then, in the role of Moderator, sums up the debate and decides the winner. (The fact that the AI plays 3 roles (3 persons, one AI) is not an issue since an AI has no ego and thus can do all 3 jobs impartially and with equal facility — cf. AlphaGo Zero, which learned Go by playing millions of games against itself and thus figuring out which patterns worked best.)
A slant razor is better than a conventional razor
This morning I had Opinionate debate this proposition: “A slant double-edge razor delivers a better shave than a conventional double-edge razor.” You can download a PDF of the full debate (or, indeed, replicate the debate for yourself). This is a low-stakes debate, particularly for those who do not shave, but I have a strong opinion about the matter. (See next post for details.)
I tried also a debate with the proposition “The shave from a slant double-edge razor is no better than the shave from a conventional razor.” Despite starting from the contradictory proposition, Opinionate reached the same conclusion, which is reassuring.
The NFL is a destructive force
It occurred to me to present Opinionate.io with propositions on various currently contentious issues to get its take on the issues. For example, submit the proposition that “NFL is a destructive force, sacrificing young men’s health to entertain the masses” to Opinionate. Opinionate in its Moderator persona sums up the debate between Debater A and Debater B. The Moderator’s conclusion:
Moderator [NFL debate]
Both debaters made strong arguments, but Debater A ultimately presented more compelling evidence supporting the premise that the NFL is a destructive force that places young men’s health at risk for the sake of entertainment. While Debater B made valid points about the choices made by players and the entertainment value of the NFL, the evidence for long-term consequences to players’ health cannot be ignored. Therefore, our winner today is Debater A.
(Unfortunately, the AI is not so good at spelling. “Debater” is the correct spelling: “debator” (probably based by analogy on “moderator”) is incorrect. I have corrected the spelling in the quoted passages.)
A whole-food plant-based diet is best
I used “whole-food plant-based diet” as a debate prompt. Here’s the Moderator’s decision:
Moderator [WFPB diet debate]
While both debaters presented strong arguments, it is clear that Debater A provided a more well-rounded and evidence-based defense of the whole-food plant-based diet. Debater A was able to effectively address Debater B’s concerns while still advocating for the benefits of a plant-based diet. Therefore, the winner of this debate is Debater A.
Debater A supported the whole-food plant-based diet. You can download the full debate as a PDF.
Firearm ownership should be regulated
I had Opinionate debate a proposition that currently divides the country: “Firearm ownership in the US should be regulated.”
Moderator [Gun regulations debate]
Both Debater A and Debater B have presented compelling arguments on the topic of firearm ownership in the US. However, based on their use of evidence and the strength of their reasoning, I would say that Debater A has provided a more nuanced and well-supported argument. While Debater B is correct that the Second Amendment protects the right to bear arms, Debater A has correctly pointed out that this right is not absolute, and that sensible regulations can be put in place to protect public safety. Additionally, Debater A has used several examples of successful gun regulations in other countries, and pointed out that freedom of speech is already restricted in some cases to ensure public safety. Therefore, I would award the win to Debater A.
Seems good to me. I agree with the moderator in all three instances.
I wonder what the impact would be if people start using propositions from their own ideology, or from an ideology they oppose) as the prompt. Would they accept an argument from an impartial AI that does not judge them?
I had Opinionate do some more debates on issues of interest to me.
A liberal-arts undergraduate education is best for a fulfilling life
A debate on the proposition: “A humanities major is the best undergraduate education for a fulfilling life.” The Moderator’s summary:
Both debaters provided strong arguments, but my ultimate decision is with Debater A. Their points that a humanities education develops critical thinking, communication, and personal growth are persuasive. While STEM fields provide financial stability and practical skills, a humanities education can give individuals the tools to make meaningful contributions to society and find fulfillment in personal and professional lives. Therefore, Debater A wins this debate.
Free education and free healthcare
A debate on the proposition “Providing free education and free healthcare to its citizens makes a nation stronger and more secure.” The Moderator’s summary:
After a thorough and respectful debate, I believe Debater A has made the stronger argument. The government’s investment in providing free education and healthcare is essential to a strong and secure nation. Investment in education and healthcare leads to a more informed and healthier citizenry, which in turn positively contributes to the economy and society as a whole. While Debater B made some valid points about the potential financial burden, the benefits of these services outweigh the negatives. Therefore, Debater A is the clear winner of this debate.
God exists only as a cultural construct
The proposition that God exists only as a cultural construct seems self-evident to me, but I was interested to see how Opinionate handled it (PDF).
And here’s a surprise: Run a debate on this proposition: “Taoism is the true religion.”
.
Schools should be allowed to show art with nudity to sixth-grade students without notifying parents in advance
Update: An example from real life.
Sidebar: Note this warning about AI
This column carries a strong warning about this new emergence of AI capability. From the column:
Social media was the first contact between A.I. and humanity, and humanity lost. First contact has given us the bitter taste of things to come. In social media, primitive A.I. was used not to create content but to curate user-generated content. The A.I. behind our news feeds is still choosing which words, sounds and images reach our retinas and eardrums, based on selecting those that will get the most virality, the most reaction and the most engagement.
While very primitive, the A.I. behind social media was sufficient to create a curtain of illusions that increased societal polarization, undermined our mental health and unraveled democracy. Millions of people have confused these illusions with reality. The United States has the best information technology in history, yet U.S. citizens can no longer agree on who won elections. Though everyone is by now aware of the downside of social media, it hasn’t been addressed because too many of our social, economic and political institutions have become entangled with it.
Large language models are our second contact with A.I. We cannot afford to lose again. But on what basis should we believe humanity is capable of aligning these new forms of A.I. to our benefit? If we continue with business as usual, the new A.I. capacities will again be used to gain profit and power, even if it inadvertently destroys the foundations of our society.
Later On 