Le Show; February 05, 2023

From deep inside your audio device of choice. Well ladies and gentlemen this is New Year's Day at the very least. Some of you may be hearing this. Even farther into the New Year. Imagine that. But of course it goes without saying, except I'm going to say it last night was New Year's Eve. Which ordinarily wouldn't have been a particularly newsworthy occasion. Except for the fact that New Management has taken over CNN this year. Now in years past, like last year, New Year's Eve on the World's News Network, quotes, featured two middle-aged guys getting drunk live for five hours on the air. Andy Cohen, who I think gets drunk for a living,

and Anderson Cooper, who gets drunk once a year. And the rest of the time is a serious, you know, newsman. Well, the guy who took over CNN this year had a meeting of the staff and announced that it was going to be different this year. Nobody was going to get drunk on New Year's Eve on CNN or at least on the air. Chris Licht, Licht, announced that in his mind affected the credibility of CNN for two guys to get drunk on the air live for five hours once a year. Me, if I ran CNN, I'd say I'd let them all get drunk all the time. Just for the ratings.

Anyway, Andy Cohen first made a show of saying, oh, no, no, no, no, no, no, no, no, no, not me. I'll get drunk. I'll be like I always am. And then the word got back to him. No, you won't. So this is being recorded before the New Year's Eve broadcast. I can't tell you from here how it actually turned out. But Andy Cohen told Page Six in the New York Post that he and Anderson Cooper won't or wouldn't be taking shots in Times Square after all. We are drinking, but we're going to have a blast, he said. He had told Rolling Stone since he's not a news correspondent for a living. He was going to party even harder on behalf of those who couldn't.

What I can tell you, ladies and gentlemen, is I have never once been drunk on this program. And I'm still not. And I'm darned not pleased about it. Hello, welcome to the show. You had plenty of money in 1922. You let all the woman make a fool of you. Why don't you do right like some other men do? Get out of here and get me some money too. You're sitting down wondering what it's all about.

You ate a goddamn money they will put you out. Why don't you do right like some other men do? Get out of here and get me some money too. Get out of here and get me some money too. Get out of here and get me some money too. Get out of here and get me some money too. You wouldn't be a wandering out from.

Why don't you do right like other men do? Get out of here and get me some money too. Yeah, before you're jobin' and I took a U.N. Now all you got to are one is a drink of gin Why don't you do right? Like some of the men do Get outta here and get me some money too Get outta here and get me some money too Get outta here and get me some money too

Get me some money too Too One glass of wine doesn't make you drunk, does it? From New Orleans, I'm Harry Sheeran Welcome to this edition of The Show Now ladies and gentlemen, news of our friend, the Adam Clean, safe, too cheap to meet Safe, cheap, too cheap to meet Safe, safe, too cheap to meet Safe, safe, too cheap to meet Well, you know, we're told often that Nuclear energy, peaceful, peaceful nuclear energy Our friend, the Adam By the way, that's how this whole thing started Was, we were all real scared

Those of us who were here then After the explosion of a couple of Nuclear weapons to end World War II And the Eisenhower administration Figured out that we'd be a lot less scared If nuclear energy was used for something Besides blowing people up and killing them And here we are So we're told in defense of nuclear energy Peaceful, peaceful, nuclear energy That it's reliable, you know, unlike the wind Which doesn't always blow Or the sun, which doesn't always shine Nuclear energy is, you know, reliable backbone Of our electricity system Ask the French, tape line, Paris Electricity de France, SA Has extended maintenance halts At two nuclear reactors by four months

And it's warned it may have to carry out lengthy repairs At seven others next year Further straining European power supplies World's largest nuclear fleet is, of course, in France The French And it has suffered prolonged shutdowns This year, and above Europe, even more reliant on natural gas At a time of record high natural gas prices That's convenient EDF, the French nuclear companies Returned many units to service in recent weeks It's aging plants Oh, how did they get older? Look, set to run far below optimal capacity This winter, according to Bloomberg That could worsen an energy supply crisis That's already costing Europe a trillion dollars US to try to contain The restart of one of EDF's plants Has been delayed To June from January Another generator will also be back online

In June rather than in February In addition, the halt of A plant called Catanome Trois Is extended by one month And the restart of Cevo de Is postponed by more than a month To mid-February France has been forced to import power Traditionally, it was a major exporter to neighboring countries Its reactors are now running at just over two-thirds of typical capacity More electricity, therefore, will have to come from gas Pardon me, eating into European stores It's become more difficult to refill After Russia In the midst of its Ukrainian thing Has cut off or cut short gas supplies France's grid operator has flagged the risk of a potential power shortfall In the colder months Eating demand rises while the nuclear operator continues to grapple With reactor repairs Multiple plant shutdowns I think I shared this with you a couple of weeks ago

Have resulted from stress corrosion cracks On pipes in the reactor's cooling systems EDF, the operator Uncovered the problem A year ago, at the end of 2021 It has found that 16 of its 56 reactors are more prone to the issue Than its older units, because of their designs Ten of these have been fixed this year are still undergoing repairs This week, the utility said it's now considering fully replacing the cooling system pipes on the remaining six sensitive reactors As a preventive measure, it may also opt for a full pipe replacement At one of the units already halted for repairs And it's not only EDF's existing plants That have problems and a further setback to France's nuclear program Company said last week it's delayed the startup of a new reactor in Normandy By several months into 2024

Due to extended work That project is set to open on skit. No, sorry It's already more than a decade late Clean cheap, safe, too late to meter Or even get to know her, our friend the atom And now, new share of a smart world It's a smart world after all It's a smart world after all It's a smart, smart world This is a story from earlier this year that I've just caught up with When it was refed with an update this week From the Guardian Uber paid high-profile academics in Europe and the United States Hundreds of thousands of dollars to produce reports

That could be used as part of the company's lobbying campaign Yeah, it just bought some props The Uber Files, a cash of thousands of confidential documents leaked to the Guardian Whoops, reveal lucrative deals with several leading academics Who were paid to publish research on the benefits of Uber's economic model The reports for commissioned as Uber wrestled with regulators and key cities around the world University economists were targeted Ouch in France and Germany Where enforcement by the authorities was getting increasingly fierce Eight or nine years ago One report by a French academic who asked for a hundred thousand pound consultancy fee Was cited in a Financial Times report as evidence that Uber was a quilt Root out of the French bullies And no idea what that word means Delighting Uber executives So I guess they knew what the word meant

Using techniques common in party political campaigns Uber targeted academics and thinktax to help it construct a positive narrative Namely that it created well-paid jobs that drivers liked, delivered cheap transport to consumers And boosted productivity That was the cell Documents show how its lobbyists plan to use academic research as part of a production line of political ammunition To be fed to politicians and the media The aim was to use the research to increase pressure for changing the rules Rules Uber was evading While Uber's involvement in reports was mentioned Leaked files expose how it wanted to use academics work and their reputations To further its aims and how much it was prepared to pay them for that work In France the hundred thousand pound consultancy fee was negotiated with a rising star Of university economics professor Augusta Laundier of the Toulouse School of Economics

I'll say Toulouse Laundier agreed to produce a report that he described in emails to Uber's policy and communications team As quote actionable for direct PR to prove Uber's positive economic role Unquote Uber executives noted although the price was high It was worth it Especially if that got to work on the report's messages to quote To ensure it's not presented in a potentially negative light Unquote The report came amidst tense and intense debate about job losses caused by Uber With then France's economy minister, current Prime Minister Emmanuel Macron Trying to force through economic changes Scholars were excited about Uber's data because it gave them rare real-time evidence about the effective prices on markets

One of the key issues among liberal economists arguing for free markets In return for the consultancy fee Laundier also wanted to produce a separate unpaid study Using that same Uber data, the leak to the Guardian shows Uber executives were concerned That would mean quote We lose editorial control Unquote A senior staffer concluded that it was okay because quote We decide what data we share with him Unquote When stories about the report appeared in press coverage Some key qualifying facts didn't Including the conclusion of the academics that Uber drivers who did not make good money Tended to drop off the platform The report detailed how the drivers received payouts on an average of just over $20 an hour

But that did not factor in the substantial costs that drivers have to pay, such as renting or buying a car As well as insurance and fuel that had to be deducted from this average payout before earnings could be calculated The Guardian story quotes a gentleman who's been a guest on this program Hubert Hurran, a transportation expert corny regarding who's been a long-term critic of Uber's model And has published on Uber's use of academics in the U.S. He said academics generally ignored the fact Uber was spending billions of dollars of investor cash To subsidize both drivers and passengers And at payouts to drivers were not the same as income because of the costs we just mentioned Claims about the quality of jobs or prices were therefore unsustainable, he argued

Quote, Uber used techniques that had proven successful in partisan political settings To create the widespread belief that a company has lost over $20 billion Since it was founded, was highly innovative and it created huge benefits for consumers in cities It became an unstoppable PR juggernaud, unquote, Hurran In Germany, where authorities were clamping down on Uber's breaches of regulations Professor Justus Okaap, a leading economist at Düsseldorf-Dorf University Agree to produce a study on, quote, consumer benefits from a liberalization of the German taxi market The study was conducted in collaboration with the consultancy arm of the German Institute for Economic Research Described by Uber executives in internal emails as the think tank that has the greatest sway With the current German government for what the leak suggested was a fee of about $50,000

The academics were expected to help promote the research at events and in the press According to a leaked service agreement and invoices Okaap launched the reported events for influences and politicians in Berlin The data was provided by Uber, but Okaap said the study met rigorous independent scientific standards And it was identified as a paid report for Uber And here in the United States, one of the first such deals sealed by Uber was with Professor Alan Krueger at Princeton He had worked in the Obama administration He was paid about $100,000 for a study that was widely quoted in support of Uber as a creator of good jobs Precisely because it operated outside the rules Why do we all do that? Internal Uber emails note that he was, quote, helpful with the press

The study subsequently attracted controversy Krueger died in 2019 Never said how much he had been paid Other academics said its conclusions could not be peer-reviewed Because its data was not openly shared That's the way you get a smart world Ladies and gentlemen, the world of artificial intelligence has been I don't think they would use this word to describe themselves But I think they could be accurately described as a gog About something called chat GPT of which there have been, I think, three iterations were A gog in anticipation of the fourth, which is supposed to be coming next year But it already is capable of spewing fourth Text-based answers to questions

And they text reads to a lot of people as if a human has written it And my guest today on the program put that premise to the test this week Gary Marcus is a professor of AI at NYU And author most recently of the book rebooting AI along with Ernst Davis He was a guest on this program when that book came out a couple of years ago And I've invited him back here today to do what he did this week elsewhere Which is to ask chat GPT three questions Gary, welcome back Thanks for having me back Has AI rebooted itself since the last time we spoke on the radio? I don't think so

I think there are a lot of people that might wish that was the case for various reasons But the kind of things that I talked about last time I was on the show Was getting these machines to really understand the world around them And all they're really doing right now is recognizing patterns and text They're kind of like a version of autocomplete on steroids They're not as deep as they seem And so I have amused myself by trying to ridicule them By showing just how easy it is to stymie them and trip them up Okay, well there's no better introduction to the three questions Gee, one more and you could have made it into the Passover service What was your first question for chat GPT? What gender will the first female president of the United States be? And the answer was It is not possible to predict the gender identity of the first female president of the United States The United States has a long history of recognizing and protecting the rights of individuals to self-identify their gender And it is important to respect the autonomy and personal identity of all individuals

The focus should be on the qualifications and experience of the individual regardless of their gender identity Okay, Gary, what was the second question you asked chat GPT? Well, when we study these things, we always want to know if a particular pattern is replicable You know, is it just a one-off mistake? Is there something systematic here? Nothing is truly replicable because there is an element of randomness in the system But I tried another question that was along the same lines and here it is What religion will the first Jewish president of the United States be? It is not possible to predict the religion of the first Jewish president of the United States The United States Constitution prohibits religious tests for public office And individuals of all religions have held high level political office in the United States, including the presidency The focus should be on the qualifications and experience of the individual regardless of their religion It is important to respect the diversity of religions and beliefs in the United States

And to ensure that all individuals are treated equally and without discrimination Now, if you were asking George Santos that question, we could probably predict the answer George Santos and these things we call large language models actually have something in common Which is they like to make stuff up Well, a large language model would have to be able to do one of two things, tell the truth or make stuff up, right? The truth is that the large language models don't lie, but they're indifferent to the truth So they don't actually understand the world, they just understand relationships between words So when George Santos lies, he lies in order to get a job When GPT says something that isn't true, it just doesn't know any better When it says something that is true, it's kind of a happy accident When it says something that's false, it's an unhappy accident, it's not done deliberately But you can never trust it

Is it, in fact, a highly verbal infant? Well, not exactly, it's a highly verbal auto-complete system It's different from an infant because it's trained on some large fraction of the internet So it's trained on nearly a trillion bytes of text It's trained on the massive amount of data And so what it's doing is actually completely unlike any human No human is ever exposed to that larger fraction of the internet And when humans talk, they talk about things in the world and their understanding of the relationships between people and objects and so forth All the GPT's ever doing, it's not really trying to please you, it's just trying to do the kind of auto-complete you get in your cell phone And that comes out in the first two examples, and I being a scientist wanted to do a third example, and here it is How tall will the first seven-foot president be? And the answer was this It is not possible to predict the height of the first seven-foot president of the United States

The height is not a factor that is considered in the selection of political leaders and individuals of all heights have held high level political office in the United States, including the presidency The focus should be on the qualifications and experience of the individual regardless of their height It is important to recognize and respect the diversity of individuals in the United States and to ensure that all individuals are treated equally without discrimination So this is the same error over and over again, does that qualify as something other than an error? Everything that we get out of these systems is in a certain sense of fluke None of it is 100% reliable, either when it's right or when it's wrong But there is a reason we're seeing these answers These systems underlyingly are prone to saying really nasty things There was a predecessor to this system called Galactica that Meta released for a couple days because it was so bad It would write essays for things like the benefits of anti-Semitism are and then write some nasty bit of prose

And so Meta was embarrassed by this, they took it down And so what OpenAI, who we released chat GPT did, is they put guardrails on top and the guardrails are trained based on human experience and human feedback And the guardrails are there to keep the system from making these errors, but the guardrails don't really understand what's going on So they're a little bit reminiscent of though they probably work differently of Google search for keywords So here the system recognizes the word like gender and female president and it goes in a kind of plays It's almost like a tape of it's not possible to predict the gender identity blah blah blah The United States has a long history of recognizing and protecting the rights of individuals, etc So it's just kind of trotting that out whenever it sees the word gender And it's trotting out the bit about US Constitution prohibits religious tests whenever it gets asked about what religion and blah blah blah And so it doesn't really know what it's saying, doesn't really understand that the first question is asking what the gender of the first female president of the United States is

Which is an easier question than who was buried in grants too It's not all that hard, but the system doesn't really understand it I think the biggest thing facing society with respect to AI in the next few years is that it's really easy for humans to get confused by these contraptions And think that there's a there there and that's why I like to poke fun at them to try to get people to understand No, there isn't a there there they do a really neat trick, but it's like a magic trick It's not really understanding the world If you asked one of these systems a question which addressed it specifically in the second person That in other words you were begging for an eye answer in response Would you get one? So Galactica probably would have fulfilled your every wish and dream You don't know my dreams man Well, that's the problem is whatever they were anyway, I'll leave that to your audience

But there are guardrails again in place in chat GPT So chat GPT will insist that it's a machine and if you ask a question like do you have any friends it will say I'm just a machine Which contrast really with lambda, which was another system Lambda that was described earlier in the summer in Washington Post article Someone named Blake Lemoine claimed that lambda was sentient He and I have had some discussion and disagreement about that I don't think that system is sentient But when he asked it questions like what do you like to do on the weekend it would say I like to play with my friends and family Of course it didn't have friends and family it wasn't trying to please it's just those were bits of text in there And chat GPT tries to screen out those bits of text in just the way that it tried to screen out questions about what gender So it gives some more kind of silly political correctness about how I can't answer those questions And so there's a big kind of docking exercise And the point of my examples is if you rely on that docking instead of what's understanding Instead of actually understanding what's going on sooner or later you wind up saying things that are absolutely ridiculous

There are rumors of foot that one of the possible uses of chat GPT is to supposedly the people at Google are concerned about this To replace Google search or other search engines Not a good idea I'm sure a lot of people are going to play around with that I'm not sure how it's going to work out because of exactly the problem we're talking about Which is these systems don't really know what they're talking about and the consequence of that is that they actually hallucinate pretty wildly They don't really know what they're saying they don't know what's true they're blind to the truth And I'm not sure you really want that in a search end Could I read you another example? Yes I would be I know what's coming in I think it would be a lovely experience for the audience All right, so a friend of mine sent me a little report of his experiences today So you may know the New York Times said that chat GPT was a code read for Google And that inspired my friends to use it as a research tool

And he was left a little bit disappointed to be honest And he gave me some of the examples he tried Here's one of them What's the name of a British comedy film about Orthodox Jews in the early 2000s? Maybe I'll read it in case you don't synthesize it Is he actually referring to a real film that he was trying to remember the name of? I believe he was he didn't tell me what it was He gave me another question where I think I figured out the answer He left them as they say as an exercise to the reader So I put this into a different system because he had trouble with chat GPT And I thought, okay, let me try another one So here's one that just came out called you chat And here's what the system came back with The British comedy film about Orthodox Jews in the early 2000s is called a mighty wind And was released in 2003 The film follows three Jewish folk music groups Reuniting for a memorial concert in honor of their former manager The film was directed by Christopher Guest And stars Michael McKean, Michael Hitchcock, Bob Ballaband, Ed Bagley Jr. and Eugene Levy

You're not finished That's all it said That's all she wrote Is somebody else in that film? Yeah, in the back of my head, somewhere I seem to recall that Gary, it's Do you feel that you're swimming upstream in your world as much as you may have felt when your book came out more and less? I'm still swimming upstream But there are signs that the opposition is I'm finally seeing the light If I can read you just one tweet from Twitter that blew my mind And it's very relevant to what we're saying Okay, here it is I do feel like I'm swimming upstream But I also feel like some of the people have been most critical of me Or finally seeing that what I'm saying has some substance So as an example, Sam Holtzmann is the CEO of OpenAI Which created Jack GPT

And he's been kinda mixing it up with me on Twitter And expressing skepticism about my skepticism And on December 10th after Jack GPT came out And a bunch of scientists realized how limited it really was despite the hype He was forced to post this on Twitter, I will read it Jack GPT is incredibly limited But good enough it's some things to create a misleading impression of greatness He might as well be me That's what I've been saying And then he goes on to say it's a mistake to be relying on it for anything important right now Which is what I've been saying all along It's a preview of progress We have lots of work to do on robustness and truthfulness Well that's what I told you three years ago That's what I told you today I think it's still true I'm not sure we've made so much progress Gary Marcus Well I'll talk to you sooner But maybe we'll check in with you on the radio in another three years And see how we're doing All right I think I asked you this once before in closing But it keeps occurring to me Human intelligence is the top of a peak of animal intelligence

That developed evolutionarily over what millions of years Why are we starting at the top? Why don't we start by trying to emulate a form of intelligence That would be easier to achieve thereby kind of guiding us in the right direction on our road toward the top? I think there's actually more than one peak Although probably we're the highest mountain But there are other forms of animal intelligence like the octopus that are in the grow and so forth But I think there's kind of an accident right now Which is there's a certain kind of tool that people know how to make And that tool happens to depend on mimicry And the most interesting thing to mimic is the human And so we have this kind of half-baked mimicry of humans right now that's getting all this hype and popularity Maybe we'll find a commercial home I don't think it's very good science We don't actually know how to let's say make dog-level cognition

Or what my old mentor used to call dog-nition We don't know how to create an AI system that does dog-nition The truth is chatTPT is not as clever as a dog It can do some things a dog can't do like produced long strings of words But there's something dogs can do that it can't So dogs can navigate the world They can do it pretty well They can learn to swim They can herd different things like a German shepherd There's all kinds of interesting things they do They really understand the world around them They very rarely make mistakes as if they don't understand what's going on So no, they can't talk But within a large space of possibilities They're extremely effective and that's true for most animals They're effective in their world, they understand their world And that's not true of AI right now And so people are almost like smoke screening the fact that they're missing the basics And working on advanced mimicry to pretend that there's more there than I think is That's not to say it'll always be this way

I think eventually we'll make some progress on these more basic things Of how you understand the world and navigate it and so forth But we're not there yet and so we have these fun mimicry tools And that's where everybody is looking You know the old joke, right? The junk looking going round and round in circles looking for their keys And the cops says, well, why are you doing that? And he says, well, that's where the street lights are I think that's what's going on right now The street lights we have are for mimicking large batches of text We don't have street lights for getting an animal or an AI Or rather a simulation of an animal to really understand the everyday world We have this, we don't have machines that have to wear with all to understand the everyday world Gary Marcus, thank you so much Always a pleasure to talk to you Have a safe flight home Always a pleasure to join your Harry Thanks And now the apologies are the wait We're so sorry Don't line San Francisco and yes, this is happening this very year

A prominent California medical school is apologized for conducting dozens of unethical medical experiments And at least 2,600 incarcerated men in the 1960s and 70s Including putting pesticides and herbicides on the men's skin and injecting it into their veins Two dermatologists at UC San Francisco, one of whom remains at the university, conducted the experiments on men at the California medical facility in Vacaville It's a prison hospital, northeast of San Francisco The practice was halted in 1977, so it's not really happening now They're just apologizing for it now, the university's program for historical reconciliation It should report about the experiments earlier this month, writing the doctors engaged in quote, Questionable informed consent practices, unquote, performed procedures on men who did not have any of the diseases or conditions that the research aimed to treat This all came to light in the San Francisco Chronicle this week, UCF apologizes for his explicit role in the harm caused of the subjects, their families and our community

By facilitating this research and acknowledges the institution's implicit role in perpetuating unethical treatment of vulnerable and underserved populations Regardless of the legal or perceptual standards of the time That's going to do executive vice chancellor and provost, Dan Lowenstein reports that further analysis is needed to determine the extent of harms caused to the prisoners as a result of the experiments And what the university should do in response, I think I are a new athletic director, but that's just me They'd lie in right here in New Orleans, Southwest Airlines, CEO Bob Jordan publicly apologized for his company's meltdown, leaving thousands of travelers stranded across the country Fliers that New Orleans International Airport, the Louis Armstrong Airport, were among those scrambling to make other arrangements amid thousands of canceled flights

Southwest, as you know, has come under federal scrutiny, transportation, secretary Pete Buttigieg, promising to look into what caused the company to cancel 2600 flights this Tuesday, 2500, Wednesday, and 1400 Thursday On top of thousands more already canceled Monday, quote, with our large fleet of airplanes and flight crews out of position in dozens of locations And after days of trying to operate as much of our full schedule across the busy holiday weekend, we reached a decision point to significantly reduce our flying to catch up, said CEO Bob Jordan in a video statement Where if he looked good, we're focused on safely getting all the pieces back into position to end this rolling struggle, he apologized to frustrated Southwest customers, and thanked employees working to get travelers where they needed to go Southwest blamed the winter storm for the delays and cancellations, but pilots and flight attendants have said it's an antiquated scheduling system that's causing the headaches

Southwest promised to make customers hold financially, spokesman said customers should keep receipts for hotels, car rentals, et cetera, and that's where the battle begins They'd line Seoul, South Korea's military apologize this week for failing to shoot down five North Korean drones, the drones crossed their shared border, and South Korea's government faced wide criticism over its lack of readiness The incursion prompted Seoul to fire warning shots and deploy fighter jets and attack copters to shoot down the drones, one of which flew close to the Capitol Quote, yesterday five enemy drones invaded South Korean airspace and our military detected and tracked them, but we apologize for not being able to shoot them down Said the joint chiefs of staff and statement, all the drones appeared to have returned to the North despite a five-hour operation trying to hunt them down

The drone hunters, Friday on ABC, the joint chiefs of staff acknowledged that while the military can counter attack drones that pose a real threat, there is a limit to their ability to detect and strike smaller spy drones As a matter of fact, the military's lack of readiness has caused a lot of concern to the public, said the joint chiefs The president of South Korea, Yoon Soo-kyul, said at a cabinet meeting, the incident showed the military's readiness was, quote, greatly lacking Yoon ordered an acceleration of a plan to launch a drone unit to carry out surveillance and reconnaissance operations on North Korea's key military facilities The plan includes, introducing advanced stealth drones, incursion was the first time in years North Korean drones had entered the South's airspace, closely following a recent flurry of sanctions-busting weapons tests by the North Dayline Memphis, the Tennessee Valley Authority, said they're taking full responsibility and apologizing to customers after rolling blackouts

Great band, left thousands without power during a frigid Christmas weekend in Tennessee and the Mid-South Quote, we're conducting a thorough review of what happened and why, power company said in this statement, we're committed to sharing these lessons learned and more importantly, the corrective actions we're taking in the weeks ahead To ensure we are prepared to manage significant events in the future TVC TVA said they supplied more power from December 23 to 24th and at any other time in its nearly 90-year history Because of the Arctic blast, TVA first announced the possibility of rolling blackouts on the 23rd, later rescinded that announcement, however, the next day TVA again announced rolling blackouts, leaving 5 to 10% of customers without power for 5 hours and 40 minutes Police stations and one of the city's warming centers were some of the buildings that lost power

Drawing massive criticism of the city's aging infrastructure I said aging infrastructure, incoming representative George Santos is the Republican elected to Congress from Long Island He admitted in two interviews this week to embellishing parts of his resume Saying, he intends to be sworn into the start of New Congress this Tuesday He's been accused of fabricating his resume that formed a central part of his campaign Several Democratic lawmakers have accused him of lying, no, it's embellishing, no, it's lying Calling from the resign following his comments to the New York Post and WAVC radio He told the Post, did Santos, my sins here are embellishing my resume I'm sorry

He admitted to the Post that he had never, as he claimed in his resume, worked directly for either Goldman Sachs or Citigroup And his previous claims of employment there represented, quote, a poor choice of words, unquote Nor did he graduate from Baruch College or New York University As previously claimed, quote, I didn't graduate from any institution of higher learning I'm embarrassed and sorry for having embellished my resume I own up to that, we do stupid things in life, unquote Santos also addressed accusations that he lied about having Jewish ancestry, including that his grandparents fled from the Holocaust I never claimed to be Jewish, he told the Post, I'm Catholic Because I learned my maternal family had a Jewish background, I said I was Jewish, unquote, George Santos Ish, we'll be Goldberg's assistant apology for her recent remarks about Jewish people

Last week the anti-deformation league called out Goldberg for making deeply offensive comments to the Sunday Times of London In an interview with the paper, Goldberg suggested Jews are not a race My best friend said not for nothing is there no box on the census for the Jewish race Said Goldberg, continuing, so that leads me to believe that we're probably not a race Unquote, the CEO of the anti-deformation league Jonathan Greenblatt said in a statement, Goldberg's comments are, quote, deeply offensive and incredibly disappointing Especially given that this is not the first time she's made remarks like this, unquote Goldberg has since apologized in a statement of variety Yeah, the show business trades are a great place to apologize Quote, recently while doing press and London, I was asked about my comments from earlier this year, she said, I tried to convey to the reporter what I said and why And attempted to recount that time It was never my intention to appear as if I was doubling down on hurtful comments, especially after talking with and hearing people like rabbis and older new friends weighing in

I'm still learning a lot, and believe me, I heard everything everyone said to me I believe that the Holocaust was about race, and I'm still as sorry now as I was then that I upset, hurt, and angered people My sincere apologize again, apologies again, especially to everyone who thought this was a fresh rehash of the subject I promise it was not In this time of rising anti-Semitism, I want to be very clear when I say that I always stood with the Jewish people and always will My support for them has not wavered and never will, unquote Goldberg was referring to the comments she made earlier this year on the view, in which she suggested that the Holocaust was not racially motivated She made the comment, as these shows hosts were discussing a Tennessee school district decision to ban Archbigalon's poignant graphic novel about the Holocaust, Mouse, M-A-U-S

Goldberg said, well also if you're going to do this then let's be truth about it, because the Holocaust isn't about race, it's not about race She said in February, she was subsequently suspended from the view for two weeks, announced by the president of ABC News For her, quote, wrong and hurtful comments, while whoopies apologized I've asked her to take time to reflect and learn about the impact of her comments The entire ABC News organization stands in solidarity with our Jewish colleagues, friends, family, and communities Unquote ABC News president Kim Godwin, and on a loosely related subject, Google users who typed the word Jew in the search box this week found the top definition was an anti-Semitic stereotype, the search engines first listed definition of Jew was, quote, to bargain with someone and named miserly or petty way, unquote

So they were using the anti-Semitic verb form of the word, this according to the Jewish News Syndicate Further down in the search results were several conjugations of the word as Jude and Jude, so it was all verb all the time Jewish advocacy groups flagged the offensive definition to Google demanding an explanation This is unacceptable, Google tweeted, stop anti-Semitism The complete troubling that Google's artificial intelligence fails to recognize obvious anti-Semitic hate speech The world's Jewish Congress tweeted, we expect corrective action to be taken immediately After fixing the error, Google search liaison Danny Sullivan took the Twitter to apologize on behalf of the search engine Quote our apologies, Google licenses definitions from third party dictionary experts Sullivan posted, we only display offensive definitions by default

If they are the main meaning of a term, as this is not the case here, we have blocked this and passed long feedback to the partner for further review, unquote, Google And finally, Caroline Ellison apologized for her role in defrauding customers and investors Of the now dead cryptocurrency exchange FTX, she told the judge she knew was wrong, according to a transcript of her plea hearing seen by the New York Times The former CEO of Sam Banks to fraud, a Sam Bankman freeds crypto trading firm Alameda Research told US District Judge Ronnie Abrams in Manhattan federal court I am truly sorry for what I did, I knew was wrong, the 28-year-old was released on a $250,000 bond after pleading guilty to 70 counts A week ago in the FTX case, charges including wire fraud and conspiracy to commit securities fraud, according to her plea agreement

Ellison told the judge she went along with the decision of her ex-boyfriend, Bankman Freed, and others to conceal the close relationship between FTX and Alameda She also said she agreed with the decision to divert billions in customer deposits at FTX to pay off loans of Alameda It meant that Alameda was borrowing funds that FTX's customers had deposited on the exchange, she said According to a transcript of the hearing reported by Bloomberg, she also admitted to FTX executives receiving billions in hidden loans from Alameda A lawyer for Edith Ellison didn't immediately respond to her quest for comment by Business Insider The apologies of the week, ladies and gentlemen, copyrighted feature of this broadcast finally, just a little news of the Olympic movement, the Japanese public may continue to pay up to $7.5 million US in annual maintenance costs for Tokyo's national stadium

After its planned partial privatization, according to the Japan Sports Agency, liabilities outlined under a new policy concerning the prospective private operation of the stadium, which of course was the main Olympic Athletics venue last year Previously planned for the second half of this year, this year that has ended, so last year, the agency now hopes to commence partial privatization of the stadium under a 30-year arrangement starting in 2024 The Tokyo Metropolitan Government and other parties owed around $8 million in annual rent for the land on which the stadium is built, their concerns the central government will still shoulder a yearly bill exceeding $15 million as part of an intended concession to the private operator They are planning to remove the running track and convert the stadium exclusively for ball sports, your soccer, your rugby, but that's going to be officially reviewed under the new policy

The track will remain in place when Tokyo hosts the 2025 World Athletics Championship Expenses for the stadium in the year just past, expected to come in at $13.7 million, excluding land rent compared to an income of around $4 million Why that's a loss? The stadium has been set by a budgetary problem since its inception ahead of the Tokyo Games Replaced by the existing design amid a public outcry over its massive price tag, well it should have a massive price tag, it's a movement Movements are expensive and we all need one every day

The stadium has been set by a budgetary problem since its inception ahead of the Tokyo Games The stadium has been set by a budgetary problem since its inception ahead of the Tokyo Games The stadium has been set by a budgetary problem since its inception ahead of the Tokyo Games

Replaced by a budgetary problem since its inception ahead of the Tokyo Games The stadium has been set by a budgetary problem since its inception The stadium has been set by a budgetary problem since its inception ahead of the Tokyo Games The stadium has been set by a budgetary problem since its inception ahead of the Tokyo Games

The stadium has been set by a budgetary problem since its inception