Come on through!
All are welcome! There is peace and serenity in the light!
(Just stay on topic, please…)
- Comment
- Reblog
-
Subscribe
Subscribed
Already have a WordPress.com account? Log in now.
Come on through!
All are welcome! There is peace and serenity in the light!
(Just stay on topic, please…)
Ethics Alarms 
Just wanted to advise the EA commentariat we found some classified documents in our garage this week.
Was there a corvette in the garage?
Just our beloved 2017 VW GTI with over a hundred thousand miles on it, so far.
In this digital day an age, why are people still printing data onto paper? Why isn’t all “classified” information kept on encrypted digital devices that are themselves kept secure? D.C. seems to be awash in paper. What’s wrong with these people?
I think you can draw a straight line between the people in charge all being elderly dimwits and the preference for printed paper. Try to imagine Biden or Trump or Pelosi reading documents on a tablet without getting confused and angry.
Yeah. As in Dilbert where the employees give the boss and Etch-A-Sketch rather than a computer.
Hillary’s leaked emails include a back and forth of her struggling to send a fax.
It’s a painful read.
I was born in the late 70s / early 80s and I don’t understand faxing. There’s only a narrow generation that understands faxing. Just scan and email. My old parents get that and my young colleagues get it. The faxing generation doesn’t.
It’s pure luck if I can find anything I’m looking for in my garage. I think the Corvette eats things.
Poor Joe from Scranton. One of Joe’s first fibs.
The oil cancer was already affecting his mental processes.
OMG! Don’t tell Joe or Donald, they’ll be wanting them for their collection… 🙂
Reminds me of my favorite Dennis Miller joke (whatever happened to Dennis Miller?) about things popping up all the time: “Was looking at my wedding video the other day and damned if Michael Caine wasn’t in it.”
Maybe Miller was “canceled.” Or maybe just tired of the bs…
I took my son to a live Miller performance. Our faces hurt we were smiling and laughing so much.
Miller had a good fun on talk radio, then a pretty good podcast for a while, but voluntarily removed himself from public life after the 2020 election. I think it was equal parts “fuck it, if you people are going to elect vegetables to the White House, I give up” and the expectation that he’d be “canceled” eventually anyway. His last few podcasts made it clear that he was very concerned that conservatives had lost the culture war, and that had put the country on a path to destruction. I think he quit because he didn’t want to be relentlessly negative, and didn’t have the energy left to fake optimism.
Interesting. And entirely plausible. Good for him. I wish more people in various walks of life knew when to retire.
Trump has some classified documents from his time when he was the classifying/declassifying authority in a certifiably secured (by corporate standards) chamber at his golf course.
The Left: END OF DEMOCRACY, END TRUMP NOW! ABSOLUTE TREASON!!!!!
Biden increasingly has classified documents from his time when he wasn’t the classifying/declassifying authority in seemingly any place he has stayed longer than one night.
The Left:
Now, rant time, as I don’t necessarily disagree with the author’s conclusion:
But, what is an “expert” in this situation?
At a minimum, I would think an “expert” on this would merely be any lawyer well versed enough the laws governing classification, redaction, or publishing of government information. And the only thing these experts could tell us whether or not something was legally classifiable. If found to be legally classifiable, all the experts can tell us is that, if classified, it isn’t “overclassified”. See, we know that the law actually gives a wide range of discretion to the executive authorities to classify information.
Which means any particular instance of classifying or redacting or publicizing information is almost always 100% a values assessment. When we get into the realm of “values”, we’re no longer involving “experts”. Literally anyone can weigh in based on what they believe is right or wrong given a full palette of values they bring to the table.
My bigger side point here has less to do with classification and more to do with my perennial gripe of removing the ability of citizens to formulate opinions by merely saying, “here’s what the experts tell us”. Just like with COVID- medical experts could try to determine and tell us the lethality of Covid, it’s spreadability, its impacts, potential mitigations, etc. But they could *never* tell us whether or not any particular societal approach to responding to Covid was “right” or “correct” or “best” if it didn’t line up with other values that are in the larger equation of life.
So, regarding classification of government information, other than lawyers telling us the details and ramifications of the laws governing the topic, anyone else weighing in, *even the authorities to whom the power is given*, are NOT “experts” on the topic. Frankly, I resent the idea that average citizens’ opinions are, by consequence of the statement, disallowed from saying “shut up government, we have a right to know generally whatever it is you are classifying and in many cases specifically purely on the principles our society claims to be founded on. We don’t need ‘experts’ to tell us this.”
Wish someone would critique this, push back, or affirm it.
It’s too obviously right to critique, experts say. 🙂
We have another outstanding example to toss on top of the already damning pile of immoral lies from just this one man.
“Petty political payback”
WHAT THE HELL!!
This man simply cannot stop lying to the people of the United States of America and he does so with utter impunity and complete disregard for anything that remotely resembles integrity. This has not been a secret, it’s widely known, because it has been stated by the Speaker of the House Rep. Kevin McCarthy multiple times prior to him becoming the Speaker of the House, he has been 100% clear and consistent and has used absolutely no uncertain terms as to why Rep. Schiff and Rep. Swalwell were removed from the Intelligence Committee.
Rep. Schiff is a bald-faced liar and political hack that knowingly abused his position on the Intelligence Committee for purely partisan political purposes. Rep. Swalwell’s known personal relationship with a Chinese spy is signature significant and it disqualifies him from sitting on any committee that deals with anything that is regarded as secret, the man can’t get a security clearance in the private business world outside of Congress, even the military wouldn’t give him a security clearance because the man has a known history of being compromised by foreign spies. These issues that Rep. McCarthy stated have been raised over and over again by Republicans and summarily dismissed by the previous Speaker of the House. It is significant that the previous Speaker of the House, Rep. Nancy Pelosi, allowed these two people to have and retain positions on the House Intelligence Committee.
As Jack has pointed out in his Concepts and Special Terms list, signature significance “posits that a single act can be so remarkable that it has predictive and analytical value, and should not be dismissed as statistically insignificant.”
Rep. Schiff is doing the same thing he’s been doing since 2016, he’s fabricating another false propaganda narrative, a lie, to feed to what he apparently sees as sheeple and the political left’s lapdog mainstream media complex (Pravda like) will willingly parrot his known lies.
“The political left has shown its pattern of propaganda lies within their narratives so many times since 2016 that it’s beyond me why anyone would blindly accept any narrative that the political left and their lapdog media actively push?” Steve Witherspoon 2022
In a “how not to” way, you have to admire the guy. He has taken the old joke, “Never tell the truth when a lie will do” to truly Pinocchio-esque levels. Although, if his nose grew every time he lied, I doubt the solar system could contain it, let alone the Earth.
Here is a LINK to Rep. Schiff’s actual tweet. There are many thousands of replies most of them talk about Rep. Schiff being a liar.
Of course I jumped on the bandwagon and piled on with my own version of replies…
“There’d be no climate crisis if it wasn’t for racism.”
Those ten (10) words, uttered by the always erudite Jane Fonda, in that sequence are quite possibly THE stupidest I’ve ever read…BAR NONE!
Yup, that about as stupid as one can get; however, stupid has no limit!
More white supremacy: Five black cops in Memphis allegedly beat a black guy to death.
We’re advised to brace for the riots.
Paul wrote:
Those ten (10) words, uttered by the always erudite Jane Fonda, in that sequence are quite possibly THE stupidest I’ve ever read…BAR NONE!
Meh. Julie Principle. 🙂
Submitted for your mocking disapproval:
We recommend avoiding general and often dehumanizing “the” labels such as the poor, the mentally ill, the French, the disabled, the college-educated. Instead, use wording such as people with mental illnesses. And use these descriptions only when clearly relevant. — Associated Press
This is one step beyond unethical besides being transparent nonsense. How can we continue to allow people like the AP and their stupid Manual of Style to continue to redefine the English language and its lexicon?
Something Must Be Done. Even though this is unintentionally food for lots of jokes and comedy, this problem is getting worse and worse. First they came for your pronouns, then they un-defined “woman,” started capitalizing colors based on racial components, and now they are coming for your grammatical articles.
From a piece by Bari Weiss’s wife (who also notes the AP style sheet absurdity about use of articles) about the Washington Post purportedly being for sale by Jeff Bezos: “If you owned The Washington Post—a place with a few great reporters, and then hundreds of screaming activists who hate journalism, hate each other, and hate you—wouldn’t you want to get the hell out?” That’s a pretty good description of the current day media.
From Legal Insurrection, the French Embassy apparently responded by tweeting “I guess this is us now” — Name “Embassy of Frenchness in the US”. Who knew that diplomats were allowed to have a sense of humor?
Top Secret-Sensitive Compartmentalized Information
This subject has been on all our minds for the past couple of weeks, so I’m going to throw in my two cents. First a little background. Before I retired from a 42 year career in the Air Force I was a long time holder of a TS-SCI security clearance.
It concerns me that people are so caught up in the documents when the truth is that the real issue is not the documents and where they are but the information those documents contain and why that information is in someone’s garage. Information of that level must be safeguarded to protect national security and prevent bad actors from actively and earnestly trying to steal it. That’s why you must go into a SCI Facility, sign-in, turn in your cell phone, before you ever learn OR DISCUSS such information. It’s also why the information is compartmentalized into need-to-know compartments. Most people don’t realize that it isn’t legal to even talk about the information outside of a SCI-Facility even if you are discussing the information with another TS-SCI holder. So the real question that should concern everyone is why even take a TS-SCI document out of a facility? Is it because you may be writing a book? You might as well write “Stolen From the US Government” in the title. Is it because you need to discuss it with someone who should have the information already and knows YOU shouldn’t be talking about it outside of a SCIF? To me there are only two alternatives: 1) it was an inadvertent (read careless) event. I accidentally hid it in my pants so as I sign out to leave the SCIF no one could see the bright red and white cover- sheets stapled to the front and back of the document to remind me to not steal it. Or 2) I needed to sneak it past the SCIF attendant so I could share the information with someone not authorized to have it.
Can my ethics colleagues think of any other reasons TS-SCI documents would be found in a garage?
Top Secret-Sensitive Compartmentalized Information
This subject has been on all our minds for the past couple of weeks, so I’m going to throw in my two cents. First a little background. Before I retired from a 42 year career in the Air Force I was a long time holder of a TS-SCI security clearance.
It concerns me that people are so caught up in the documents when the truth is that the real issue is not the documents and where they are but the information those documents contain and why that information is in someone’s garage. Information of that level must be safeguarded to protect national security and prevent bad actors from actively and earnestly trying to steal it. That’s why you must go into a SCI Facility, sign-in, turn in your cell phone, before you ever learn OR DISCUSS such information. It’s also why the information is compartmentalized into need-to-know compartments. Most people don’t realize that it isn’t legal to even talk about the information outside of a SCI-Facility even if you are discussing the information with another TS-SCI holder. So the real question that should concern everyone is why even take a TS-SCI document out of a facility? Is it because you may be writing a book? You might as well write “Stolen From the US Government” in the title. Is it because you need to discuss it with someone who should have the information already and knows YOU shouldn’t be talking about it outside of a SCIF? To me there are only two alternatives: 1) it was an inadvertent (read careless) event. I accidentally hid it in my pants so as I sign out to leave the SCIF no one could see the bright red and white cover- sheets stapled to the front and back of the document to remind me to not steal it. Or 2) I needed to sneak it past the SCIF attendant so I could share the information with someone not authorized to have it.
Can my ethics colleagues think of any other reasons TS-SCI documents would be found in a garage?
It is frightening to think that this vital information is so often trusted to ethically barren politicians.
It is frightening to think that this vital information is so often trusted to ethically barren politicians.
Or else Hunter was selling the Ukes classified information about Ukraine and its various gas companies. rather than merely the chance to have a photo taken with Joe.
Do we know if any of the documents found at Trump’s, Biden’s or Pence’s locations have, or continue to retain a Top Secret-Sensitive Compartmentalized Information level of security?
Do we know if any of the documents found at Trump’s, Biden’s or Pence’s locations have, or continue to retain a Top Secret-Sensitive Compartmentalized Information level of security? 😉
I’ve heard lots of innuendo but nothing concrete on that question yet.
That’s a fair question Steve. One would think that if non of the information is TS-SCI it would cool down the rhetoric to inform the public of that fact.
Kent Clark wrote, “One would think that if non of the information is TS-SCI it would cool down the rhetoric to inform the public of that fact.”
You’d think that but that’s not the narrative that the political left and their Pravda like media want to push about Trump and Republicans. It’s all about framing the propaganda narrative.
What’s true is the political left and their lapdog Pravda like media will intentionally rationalize the Biden secret documents and hide any serious related facts about those documents while they continue to demonize Pence and especially Trump’s secret documents and actively promote unsupported innuendo as to what those documents are.
Actual truth, the whole truth and nothing but the truth is not within the ideological frame work of today’s progressive political left.
“The political left has shown its pattern of propaganda lies within their narratives so many times since 2016 that it’s beyond me why anyone would blindly accept any narrative that the political left and their lapdog media actively push?” Steve Witherspoon 2022
It seems to me that Pence’s revelation that he had classified documents was intended to put an end to this mess. Pence is considered a very reasonable, prudent, serious politician. If he had them, then he is saying that document retention and classification system is way too loose and the laws governing such matters are antiquated and outdated.
jvb
“Can my ethics colleagues think of any other reasons TS-SCI documents would be found in a garage?”
To be fair, there is the possibility that the document was created on-site. Anyone with knowledge of classified information (which the Vice-President of the United States surely would have) could theoretically write any of it down anywhere, any time.
This is still a gross violation of the rules, but at least a kind-of plausible explanation . . . ?
–Dwayne
I want to posit the following question to those who advocate for equal outcomes in all areas of society.
Let’s assume that we have reached a perfect representation of population percentages of each identified demographic segments in every trade, profession, and public group. How do we assign who is to do what if the population dynamics shift or the wants and needs of society change in such a manner that some industries and jobs are created or disappear.
Obviously, this will require some form of central planning to ensure legislated “equity”. How would you democratize the planning process when every member of every group demands that they get a position of wealth and power? You cannot. Therefore, a select group will choose for the society as a whole.
Logically it seems to me that this would undermine a women’s right to choose. Such a system would require that we accept sub-standard output if people are no allowed to engage in areas that they have greater aptitude which leads to inefficiencies in production and consumption. And is it likely that a caste system would develop such that upward mobility would be limited to those on the bottom because those on the top would be able to manipulate the system so that their offspring are selected to be in the higher social caste?
I’m still waiting for NBA and NFL teams to be limited to no more than thirteen percent of rosters being black guys. Or, come to think of it, shouldn’t half of the players be women?
Or Jews in any endeavor to be limited to three percent. Hollywood, Wall Street, sports franchise ownership, you name it.
It always hinges on choice for me but not for thee.
I really don’t think these equity hawks understand how equilibrium is achieved in consumption and production. There will always be some inefficiencies due to asymmetric information but when we try to force an equilibrium we find that unseen forces create pressure that will ultimately need to be relieved and the longer that growing pressure is contained the more uncontrolled kinetic energy will be released when the containment vessel ruptured.
As promised, my thought on the current round of GPT-based Artificial Intelligence systems. First, some background. I’m an EE, graduated in 2004, but focused most of my career on software development. Wrote software professionally for one of the big tech companies for more than 12 years before going back to my roots as a hardware-level developer. For the last two years I’ve been doing avionics software, but my current employer’s flight stack includes a number of AI components for autonomous flight with which I interact regularly.
So let’s talk about the GPT family of models. These are language models, in the sense that what they do is take a sequence of text and predict what the logical continuation would be. In order to achive this they are fed a large amount of data to base their predictions on. This includes decades of newspaper and magazine articles, scripts for plays and movies, technical journal articles, chat transcripts, etc. The selection and filtering of data is key (otherwise it would learn spelling and grammatical errors). From what I have seen so far the data that is fed to the model is fairly good quality.
The system then builds an extremely complex prediction model – this can be measured by what is vaguely referred to as “parameters” – each just being a number in a large algorithm (think recipe or math formula) where you feed some text as an input and get as an output some more text that be a continuation of the provided input. Different models are optimized for different types of text. Generic GPT would usually generate news report style pieces: somewhat short and with easy to comprehend vocabulary and grammar. ChatGPT is tailored instead to generate the next response to a dialog. It will take as an input the previous statements in the conversation and create the response that would follow logically.
This is the point where I am going to get a little technical.
To emphasize: the ChatGPT model generates its response based only on the provided text by the user and its previous responses. Every time you respond to a conversation it has to replay the whole conversation and then predicts in some way what the appropriate reply is. In this system there is no internal state to keep track of. This makes it extremely easy to have it say nonsensical things. An example:
Me: What letter comes after E?
ChatGPT: F
Me: You are incorrect, the letter that comes after E is G.
ChatGPT: I apologize, you are correct. The letter that comes after E is G.
(This is a verbatim transcript of a chat I had with ChatGPT).
I start to suspect it has no real model of what the alphabet is. So I follow up with:
Me: What are letters 5 through 7 of the alphabet?
ChatGPT: F, G, H [editor’s note: this is off by one]
Me: What are the letters 1 though 7 of the alphabet?
ChatGPT: A, B, C, D, E, F, G
At this point a system with any internal awareness would recognize that one mistake (at least) was made. Oh well, maybe ChatGPT will notice if I point it out.
Me: Is there a contradiction in your two previous responses?
ChatGPT: No, there is no contradiction in my previous responses. The first response is answering a question about the letter that comes after E and the second response is answering a question about the letters 1 through 7 of the alphabet which starts from A.
Well, now it’s clear there is no “there” there. It does not understand or reason about the conversation. And while this should not come as a surprise once one understands how the language model is build and works, seeing it clearly illustrated can be striking – specially if one follows the hype by the tech media and all the chicken little luddites about AI doing everything from taking our jobs to destroying the world.
On the other hand, I will grant that as a language model it is very good. The text it generates is syntactically correct, easy to read, and has no typos (wink). Someone looking superficially at it could conclude that now we have a writer that can pass for human. That would be bad. I’m guessing most of the readers here would not be fooled and identify the errors and lack of substance after reading a few paragraphs. But there is a significant portion of the population that might be fooled by pretty-looking text, even if the content is obviously incorrect on more careful inspection.
And this is where the ethics question begins. What are the consequences of releasing such a tool? What if a propagandist can now write thousands and thousands of pages of text to overwhelm the discourse (instead of just buying a major newspaper)? What happens when AI-created content is so much that it becomes impossible to find the actual human insights? Can we prevent the public from being fooled by these artifical but slick salesmen?
I do not have answers for everything, but I have some thoughts.
First, AI – or “AI” because once you understand how it works it’s clear there is no intelligence – is nowhere near a point where it can replace knowledge workers except at the lowest tiers. Maybe your offshore call center can be replaced with a server farm, after all those guys are not allowed to deviate from the script either (they won’t even tell you the answer of 2+2 if you ask, I use that as a test to see if they have the minimum authority to solve my problem). Anything that requires one iota of creativity or initiative is perfectly safe.
Second, as in most cases, the tools themselves are not ethical or unethical, it’s their use that will be. We will need to prepare ourselves for dealing with the outcomes, but this is not the first time we’ve had to deal with a broken tech (see spam, bad search results, automated robocalls). They are annoyances, and they make the world a little worse, but they are not going to break down society (any more than it is already broken).
Third, given that these approaches give quick and flashy “wins”, actual AI research based on creating generalized understanding models is receiving almost no funding, and very little interest. Back in 2006-07 I worked on a team that was figuring out models of semantic understanding to allow paraplegic users to control a computer efficiently, instead of having to use alternative inputs that ended up looking like mouse movements and keyboard keystrokes. It is a very hard problem, but also very interesting. You need to self analyze your own thought processes to extract intent, then figure out the steps for solving the problem. Instead, the first attempts at statistics-based systems (like the ones currently in vogue) gave faster results and teams like mine were defunded. We have lost almost two decades of advancement that we will not be getting back.
Fourth, as individuals the best one can do is learn to identify “AI”-generated content and reject it. As a society I do not have an answer, because there are people who are unwilling or unable to develop this skill.
Happy to chat more about any of this!
I know it’s a little late, but I have a day job. 🙂
Excellent comment and quite interesting. It’s actually a bit discouraging that research on genuine AI is being shorted, but I’m also not altogether sure that is necessarily a bad thing.
You make me hearken back to my youth when programmers were just starting to write programs to play chess — back then the computer programs were ‘fish’, but they have solved the chess problem. From what you’re saying, I do wonder just how heuristically programmers have solved the problem. In theory, chess is a deterministic game — the problem was always having enough computational power to solve it that way.
I know I regularly get phone calls that pretend to be from humans — in my experience it is not hard and doesn’t take long to determine when they are, in fact, robocalls. Then again, if I am in the mood, with the actual human callers I can happily waste five or ten minutes of their time that they could otherwise be using to scam someone. I have a good time, they’ll never get those ten minutes of their life back — it’s a win, win in my book.
Re: discussion of sexuality or orientation in classrooms.
If it is inappropriate for a manager to discuss sexual issues in to workplace because it could be deemed sexual harassment due to an inherent power imbalance, how can it be reconciled, that given the very same power imbalance that exists between student and teacher, discussions about the teacher’s own sexual preferences with subordinate children is acceptable. It seems to me that if adult discussions of a sexual nature in the workplace can lead to harassment suits and big payouts then discussions with children is even more inappropriate.
Why has no one even made this comparison? What argument could be used to justify the latter?
Something to consider.
https://reason.com/2023/01/27/twitter-files-matt-taibbi-hamilton-68-russian-bots-fake/
Yeah, I can’t keep up with the Twitter Files. I had no idea they would be this extensive….or damning.
How about the guy supposedly investigating Trump’s Russian connections is now accused of . . . get this . . . colluding with the . . . wait for it . . . RUSSIANS!!!! It just keeps getting stranger and stranger.
jvb
It’s clear to me now that we’re living in an artificially constructed reality, very similar to “The Matrix”. The big difference from the movie however, is that the artificial intelligence that controls everything is not using us as “batteries”, harvesting energy from our bodies. The purpose appears to be simply to fuck with us for its own amusement. How else to explain the utterly ridiculous twists and turns of the last few years?