Connect with us

Sci-Tech

The New ChatGPT Offers a Lesson in AI Hype

Published

on


When OpenAI unveiled the latest version of its immensely popular ChatGPT chatbot this month, it had a new voice possessing humanlike inflections and emotions. The online demonstration also featured the bot tutoring a child on solving a geometry problem.

To my chagrin, the demo turned out to be essentially a bait and switch. The new ChatGPT was released without most of its new features, including the improved voice (which the company told me it postponed to make fixes). The ability to use a phone’s video camera to get real-time analysis of something like a math problem isn’t available yet, either.

Amid the delay, the company also deactivated the ChatGPT voice that some said sounded like the actress Scarlett Johansson, after she threatened legal action, replacing it with a different female voice.

For now, what has actually been rolled out in the new ChatGPT is the ability to upload photos for the bot to analyze. Users can generally expect quicker, more lucid responses. The bot can also do real-time language translations, but ChatGPT will respond in its older, machine-like voice.

Nonetheless, this is the leading chatbot that upended the tech industry, so it was worth reviewing. After trying the sped-up chatbot for two weeks, I had mixed feelings. It excelled at language translations, but it struggled with math and physics. All told, I didn’t see a meaningful improvement from the last version, ChatGPT-4. I definitely wouldn’t let it tutor my child.

This tactic, in which A.I. companies promise wild new features and deliver a half-baked product, is becoming a trend that is bound to confuse and frustrate people. The $700 Ai Pin, a talking lapel pin from the start-up Humane, which is funded by OpenAI’s chief executive, Sam Altman, was universally panned because it overheated and spat out nonsense. Meta also recently added to its apps an A.I. chatbot that did a poor job at most of its advertised tasks, like web searches for plane tickets.

Companies are releasing A.I. products in a premature state partly because they want people to use the technology to help them learn how to improve it. In the past, when companies unveiled new tech products like phones, what we were shown — features like new cameras and brighter screens — was what we were getting. With artificial intelligence, companies are giving a preview of a potential future, demonstrating technologies that are being developed and working only in limited, controlled conditions. A mature, reliable product might arrive — or might not.

The lesson to learn from all this is that we, as consumers, should resist the hype and take a slow, cautious approach to A.I. We shouldn’t be spending much cash on any underbaked tech until we see proof that the tools work as advertised.

The new version of ChatGPT, called GPT-4o (“o” as in “omni”), is now free to try on OpenAI’s website and app. Nonpaying users can make a few requests before hitting a timeout, and those who have a $20 monthly subscription can ask the bot a larger number of questions.

OpenAI said its iterative approach to updating ChatGPT allowed it to gather feedback to make improvements.

“We believe it’s important to preview our advanced models to give people a glimpse of their capabilities and to help us understand their real-world applications,” the company said in a statement.

(The New York Times sued OpenAI and its partner, Microsoft, last year for using copyrighted news articles without permission to train chatbots.)

Here’s what to know about the latest version of ChatGPT.

To show off ChatGPT-4o’s new tricks, OpenAI published a video featuring Sal Khan, the chief executive of the Khan Academy, the education nonprofit, and his son, Imran. With a video camera pointed at a geometry problem, ChatGPT was able to talk Imran through solving it step by step.

Even though ChatGPT’s video-analysis feature has yet to be released, I was able to upload photos of geometry problems. ChatGPT solved some of the easier ones correctly, but it tripped up on more challenging problems.

For one problem involving intersecting triangles, which I dug up on an SAT preparation website, the bot understood the question but gave the wrong answer.

Taylor Nguyen, a high school physics teacher in Orange County, Calif., uploaded a physics problem involving a man on a swing that is commonly included on Advanced Placement Calculus tests. ChatGPT made several logical mistakes to give the wrong answer, but it was able to correct itself with feedback from Mr. Nguyen.

“I was able to coach it, but I’m a teacher,” he said. “How is a student supposed to pick out those mistakes? They’re making this assumption that the chatbot is right.”

I did notice that ChatGPT-4o succeeded at some division calculations that its predecessors did incorrectly, so there are signs of slow improvement. But it also failed at a basic math task that past versions and other chatbots, including Meta AI and Google’s Gemini, have flunked at: the ability to count. When I asked ChatGPT-4o for a four-syllable word starting with the letter “W,” it responded, “Wonderful.”

OpenAI said it was constantly working to improve its systems’ responses to complex math problems.

Mr. Khan, whose company uses OpenAI’s technology in its tutoring software Khanmigo, did not respond to a request for comment on whether he would leave ChatGPT the tutor alone with his son.

OpenAI also highlighted that the new ChatGPT was better at reasoning, or using logic to come up with responses. So I ran it through one of my favorite tests: I asked it to generate a Where’s Waldo? puzzle. When it showed an image of a giant Waldo standing in a crowd, I said that the point is that he’s supposed to be hard to find.

The bot then generated an even larger Waldo.

Subbarao Kambhampati, a professor and researcher of artificial intelligence at Arizona State University, also put the chatbot through some tests and said he saw no noticeable improvement in reasoning compared with the last version.

He presented ChatGPT a puzzle involving blocks:

If block C is on top of block A, and block B is separately on the table, can you tell me how I can make a stack of blocks with block A on top of block B and block B on top of block C, but without moving block C?

The answer is that it’s impossible to arrange the blocks under these conditions, but, just as with past versions, ChatGPT-4o consistently came up with a solution that involved moving block C. With this and other reasoning tests, ChatGPT was occasionally able to take feedback to get the correct answer, which is antithetical to how artificial intelligence is supposed to work, Mr. Kambhampati said.

“You can correct it, but when you do that you’re using your own intelligence,” he said.

OpenAI pointed to test results that showed GPT-4o scored about two percentage points higher at answering general knowledge questions than previous versions of ChatGPT, illustrating that its reasoning skills had slightly improved.

OpenAI also said the new ChatGPT could do real-time language translation, which could help you converse with someone speaking a foreign language.

I tested ChatGPT with Mandarin and Cantonese and confirmed that it was OK at translating phrases, such as “I’d like to book a hotel room for next Thursday” and “I want a king-size bed.” But the accents were slightly off. (To be fair, my broken Chinese is not much better.) OpenAI said it was still working to improve accents.

ChatGPT-4o also excelled as an editor. When I fed it paragraphs that I wrote, it was fast and effective at removing excessive words and jargon. ChatGPT’s decent performance with language translation gives me confidence that this will soon become a more useful feature.

A major thing OpenAI got right with ChatGPT-4o is making the technology free for people to try. Free is the right price: Since we are helping to train these A.I. systems with our data to improve, we shouldn’t be paying for them.

The best of A.I. has yet to come, and it might one day be a good math tutor that we want to talk to. But we should believe it when we see it — and hear it.



Source link

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Sci-Tech

As the E.V. Revolution Slows, Ferrari Enters the Race

Published

on


Gliding on robotic haulers, a line of Ferrari frames maneuvers through a gleaming new factory in Northern Italy. At each station, engineers in cherry red uniforms add a component — an engine block, a dashboard, a steering wheel — as they transform the bodies into hybrid vehicles. Up next: fully electric.

A lot is riding on Ferrari’s 200-million-euro “e-building,” which went into operation last month and is nearly twice the size of Rome’s Colosseum. The factory is intended to bring the 77-year-old sports-car maker, known for the sonorous vroom of its gas engines, into the age of electrification.

But the effort comes at a precarious time for the auto industry. The transition to electric vehicles, which was supposed to quickly usher in an era of climate-friendly transport, has instead been squeezed by costly investments and slowing global demand.

Other luxury carmakers have struggled to go electric. Mercedes-Benz and Lamborghini have reduced their ambitions. Tesla reported declining second-quarter sales on Tuesday, and Ford Motor said in April that it would shift production to more hybrids as E.V. losses piled up. A growing trade war between China and the West also threatens to stifle growth.

Despite the challenges, Ferrari sees an opportunity in the industry’s inevitable march toward electrification to reach a new consumer: the wealthy environmentalist. It intends to unveil its first fully electric model in the fourth quarter of next year. As part of its strategy, the carmaker has enlisted LoveFrom — the agency founded by Jony Ive, Apple’s former design chief, and the industrial designer Marc Newson — to hone the car’s appearance.

There is plenty of mystery shrouding the yet-to-be-named car, including its battery life and what it will sound like. The company has not disclosed its look, production run or price tag. But it could be one of the most expensive electric vehicles on the market, analyst say, surpassing Porsche’s $286,000 Taycan Turbo GT.

Ferrari’s foray into electric will be notable for other reasons. Regulators may be pushing electric vehicles, but there is lingering skepticism in the marketplace. Winning over fans of combustion engines will not be easy — even for Ferrari. And the industry is desperate for an automaker, any automaker, to prove that electric vehicles can drive big profits.

“It’s worth watching whether a Ferrari E.V. can maintain the kind of price premium you’d associate with a Ferrari,” said Martino de Ambroggi, an automotive analyst at Equita, an investment bank in Milan. “Often, a Ferrari purchase is also viewed as a kind of investment. Only after a few years will we see if that investment in an electric Ferrari holds up.”

Benedetto Vigna, Ferrari’s chief executive, is doing his best to keep the market in anticipation. In an interview last month in the new plant, he said the company would commence full-scale electric vehicle production by early 2026. By 2030, electric and hybrid cars will make up as much as 80 percent of Ferrari’s annual output as the company seeks to meet stringent European Union emissions mandates.

In the meantime, the e-building will roll out two models: the SF90 Stradale, a plug-in hybrid, and the combustion engine Purosangue.

Ferrari does not need an electric vehicle to pad its bottom line. Under Mr. Vigna, a former executive at the chip maker STMicroelectronics who took the helm nearly three years ago, the company has been on a tear. The stock is one of the best performers in Europe this year, giving it a roughly $75 billion market valuation, higher than that of Ford or General Motors. Profits are soaring alongside prices at Ferrari, which makes some of the most expensive cars on the planet. There’s a three-year waiting list for some models.

Ferrari’s success over the years on the Formula 1 track has also led to a lucrative corporate sponsorship and merchandise business that has transformed it into a luxury brand with a sporty flair. Ferrari’s prancing horse logo can be found on high-end apparel like a €790 cashmere sweater.

Mr. Vigna sees the electric vehicle as part of the company’s growth strategy, despite the industry’s slowdown. “There are some potential clients, I have them clearly in mind, who will never become part of the family unless there is an electric car,” he said.

But challenges loom. Enthusiasts who had gathered outside the factory gates last month wondered: Will it look, handle and sound like the classic Ferrari growler, or have the understated whine of most electric vehicles?

“When you think of a Ferrari, it still has that kind of engine sensation, and you also think of the roar,” Mr. de Ambroggi said. “I don’t know how Ferrari resolves this.”

Mr. Vigna fields that question often, especially from longtime customers, or Ferraristi. They seem to be channeling the deceased founder, Enzo Ferrari, who once broke down in the simplest terms how he built some of the fastest cars on the planet: “I build motors and attach them to wheels.”

Mr. Vigna’s E.V. pitch has a different ring. “The electric engine will not be silent,” he said. “There are ways to make sure that the emotion comes through from driving an electric Ferrari that is the same as when you drive a hybrid or when you drive a thermal Ferrari.”

Battery life is another puzzle piece. Because Ferraris often sell for a higher price on the secondary market, the concern about battery degradation, and its impact on the long-term value of the car, may be felt more acutely by the Ferraristi.

“The E.V. transition raises a whole lot of new issues for them in terms of how you maintain the vehicle,” said Stephen Reitman, an auto analyst at Bernstein.

Ferrari’s longtime partner, SK On, a South Korean battery maker, will supply the components for the E.V. batteries, which Ferrari will assemble in the e-building, where it will also make the car’s electric motors and axles.

And then there is the matter of price. Last month, Reuters reported that the car would cost at least €500,000 ($540,000). Mr. Vigna pushed back on the speculation, saying it is too early to talk price.

Ferrari still follows its founder’s principle for producing a limited number of extremely expensive cars. Ferrari made fewer than 14,000 last year; even with the e-building, production is not expected to increase much at the start.

The limited numbers may explain why fans make the pilgrimage to Maranello hoping to catch a glimpse of a Ferrari, either on the company’s Formula 1 test track or near its red brick factory.

Knowing demand is high, Mr. Vigna has increased the base price of most models more than 25 percent.

“Ferrari consistently sells less than the market demands, leading to a multiyear order book,” said Mr. Reitman, the Bernstein analyst. With a profit margin of nearly 30 percent, Ferrari’s business more resembles that of a luxury brand like Hermes or Rolex, analysts say.

Mr. Vigna is already thinking about how to market the new electric car. The target customer probably will not be buying the car for purely practical or even planet-saving reasons, he said, adding: “The emotional part of the brain is driving the purchase.”



Source link

Continue Reading

Sci-Tech

Students Target Teachers in Group TikTok Attack, Shaking Their School

Published

on


In February, Patrice Motz, a veteran Spanish teacher at Great Valley Middle School in Malvern, Pa., was warned by another teacher that trouble was brewing.

Some eighth graders at her public school had set up fake TikTok accounts impersonating teachers. Ms. Motz, who had never used TikTok, created an account.

She found a fake profile for @patrice.motz, which had posted a real photo of her at the beach with her husband and their young children. “Do you like to touch kids?” a text in Spanish over the family vacation photo asked. “Answer: Sí.”

In the days that followed, some 20 educators — about one quarter of the school’s faculty — discovered they were victims of fake teacher accounts rife with pedophilia innuendo, racist memes, homophobia and made-up sexual hookups among teachers. Hundreds of students soon viewed, followed or commented on the fraudulent accounts.

In the aftermath, the school district briefly suspended several students, teachers said. The principal during one lunch period chastised the eighth-grade class for its behavior.

The biggest fallout has been for teachers like Ms. Motz, who said she felt “kicked in the stomach” that students would so casually savage teachers’ families. The online harassment has left some teachers worried that social media platforms are helping to stunt the growth of empathy in students. Some teachers are now hesitant to call out pupils who act up in class. Others said it had been challenging to keep teaching.

“It was so deflating,” said Ms. Motz, who has taught at the school, in a wealthy Philadelphia suburb, for 14 years. “I can’t believe I still get up and do this every day.”

The Great Valley incident is the first known group TikTok attack of its kind by middle schoolers on their teachers in the United States. It’s a significant escalation in how middle and high school students impersonate, troll and harass educators on social media. Before this year, students largely impersonated one teacher or principal at a time.

The middle schoolers’ attack also reflects broader concerns in schools about how students’ use, and abuse, of popular online tools is intruding on the classroom. Some states and districts have recently restricted or banned student cellphone use in schools, in part to limit peer harassment and cyberbullying on Instagram, Snap, TikTok and other apps.

Now social media has helped normalize anonymous aggressive posts and memes, leading some children to weaponize them against adults.

“We didn’t have to deal with teacher-targeting at this scale before,” said Becky Pringle, president of the National Education Association, the largest U.S. teachers’ union. “It’s not only demoralizing. It could push educators to question, ‘Why would I continue in this profession if students are doing this?’”

In a statement, the Great Valley School District said it had taken steps to address “22 fictitious TikTok accounts” impersonating teachers at the middle school. It described the incident as “a gross misuse of social media that profoundly impacted our staff.”

Last month, two female students at the school publicly posted an “apology” video on a TikTok account using the name of a seventh-grade teacher as a handle. The pair, who did not disclose their names, described the impostor videos as a joke and said teachers had blown the situation out of proportion.

“We never meant for it to get this far, obviously,” one of the students said in the video. “I never wanted to get suspended.”

“Move on. Learn to joke,” the other student said about a teacher. “I am 13 years old,” she added, using an expletive for emphasis, “and you’re like 40 going on 50.”

A TikTok account displaying the name of a Great Valley Middle School teacher posted a video in late June about the student suspensions.Credit…via TikTok

In an email to The New York Times, one of the students said that the fake teacher accounts were intended as obvious jokes, but that some students had taken the impersonations too far.

A TikTok spokeswoman said the platform’s guidelines prohibit misleading behavior, including accounts that pose as real people without disclosing that they are parodies or fan accounts. TikTok said a U.S.-based security team validated ID information — such as driver’s licenses — in impersonation cases and then deleted the data.

Great Valley Middle School, known locally as a close-knit community, serves about 1,100 students in a modern brick complex surrounded by a sea of bright green sports fields.

The impostor TikToks disrupted the school’s equilibrium, according to interviews with seven Great Valley teachers, four of whom requested anonymity for privacy reasons. Some teachers already used Instagram or Facebook but not TikTok.

The morning after Ms. Motz, the Spanish teacher, discovered her impersonator, the disparaging TikToks were already an open secret among students.

“There was this undercurrent conversation throughout the hallway,” said Shawn Whitelock, a longtime social studies teacher. “I noticed a group of students holding a cellphone up in front of a teacher and saying, ‘TikTok.’”

Students took images from the school’s website, copied family photos that teachers had posted in their classrooms and found others online. They made memes by cropping, cutting and pasting photos, then superimposing text.

The low-tech “cheapfake” images differ from recent incidents in schools where students used artificial intelligence apps to generate real-looking, digitally altered images known as “deepfakes.”

While some of the Great Valley teacher impostor posts seemed jokey and benign — like “Memorize your states, students!” — other posts were sexualized. One fake teacher account posted a collaged photo with the heads of two male teachers pasted onto a man and woman partially naked in bed.

Fake teacher accounts also followed and hit on other fake teachers.

“It very much became a distraction,” Bettina Scibilia, an eighth-grade English teacher who has worked at the school for 19 years, said of the TikToks.

Students also targeted Mr. Whitelock, who was the faculty adviser for the school’s student council for years.

A fake @shawn.whitelock account posted a photo of Mr. Whitelock standing in a church during his wedding, with his wife mostly cropped out. The caption named a member of the school’s student council, implying the teacher had wed him instead. “I’m gonna touch you,” the impostor later commented.

I spent 27 years building a reputation as a teacher who is dedicated to the profession of teaching,” Mr. Whitelock said in an interview. “An impersonator assassinated my character — and slandered me and my family in the process.”

Mrs. Scibilia said a student had already posted a graphic death threat against her on TikTok earlier in the school year, which she reported to the police. The teacher impersonations increased her concern.

“Many of my students spend hours and hours and hours on TikTok, and I think it’s just desensitized them to the fact that we’re real people,” she said. “They didn’t feel what a violation this was to create these accounts and impersonate us and mock our children and mock what we love.”

A few days after learning of the videos, Edward Souders, the principal of Great Valley Middle School, emailed the parents of eighth graders, describing the impostor accounts as portraying “our teachers in a disrespectful manner.”

In early March, the principal of Great Valley Middle School, Edward Souders, sent eighth-grade parents an email about the impostor accounts on TikTok.

The school also held an eighth-grade assembly on responsible technology use.

But the school district said it had limited options to respond. Courts generally protect students’ rights to off-campus free speech, including parodying or disparaging educators online — unless the students’ posts threaten others or disrupt school.

“While we wish we could do more to hold students accountable, we are legally limited in what action we can take when students communicate off campus during nonschool hours on personal devices,” Daniel Goffredo, the district’s superintendent, said in a statement.

The district said it couldn’t comment on any disciplinary actions, to protect student privacy.

In mid-March, Nikki Salvatico, president of the Great Valley Education Association, a teachers’ union, warned the school board that the TikToks were disrupting the school’s “safe educational environment.”

“We need the message that this type of behavior is unacceptable,” Ms. Salvatico said at a school board meeting on March 18.

The next day, Dr. Souders sent another email to parents. Some posts contained “offensive content,” he wrote, adding: “I am optimistic that by addressing it together, we can prevent it from happening again.”

In mid-March, Dr. Souders, the principal, sent a second email to parents, this time noting that some of the TikToks contained “offensive content.”

While a few accounts disappeared — including those using the names of Ms. Motz, Mr. Whitelock and Mrs. Scibilia — others popped up. In May, a second TikTok account impersonating Mrs. Scibilia posted several new videos mocking her.

She and other Great Valley educators said they had reported the impostor accounts to TikTok, but had not heard back. But several teachers, who felt the videos had violated their privacy, said they did not provide TikTok with a personal ID to verify their identities.

On Wednesday, TikTok removed the account impersonating Mrs. Scibilia and three other fake Great Valley teacher accounts flagged by a reporter.

Mrs. Scibilia and other teachers are still processing the incident. Some teachers have stopped posing for and posting photographs, lest students misuse the images. Experts said this type of abuse could harm teachers’ mental health and reputations.

“That would be traumatizing to anyone,” said Susan D. McMahon, a psychology professor at DePaul University in Chicago and chair of the American Psychological Association’s Task Force on Violence Against Educators. She added that verbal student aggression against teachers was increasing.

Now teachers like Mrs. Scibilia and Ms. Motz are pushing schools to educate students on how to use tech responsibly — and bolster policies to better protect teachers.

Great Valley students on TikTok warned their schoolmates that teachers had learned of the impostor accounts.Credit…via TikTok

In the Great Valley students’ “apology” on TikTok last month, the two girls said they planned to post new videos. This time, they said, they would make the posts private so teachers couldn’t find them.

“We’re back, and we’ll be posting again,” one said. “And we are going to private all the videos at the beginning of next school year,” she added, “’cause then they can’t do anything.”

On Friday, after a Times reporter asked the school district to notify parents about this article, the students deleted the “apology” video and removed the teacher’s handle from their account. They also added a disclaimer: “Guys, we’re not acting as our teachers anymore that’s in the past !!”





Source link

Continue Reading

Sci-Tech

Mark Zuckerberg’s Viral Surf Video

Published

on


As Fourth of July celebrations commenced across the nation, Mark Zuckerberg dropped a video onto his Instagram account that immediately generated hundreds of thousands of views. Indeed, the clip seemed designed for warp speed virality.

Behind a fast-moving boat, Mr. Zuckerberg wakeboards while wearing a tuxedo and sunglasses as he sips from a tall boy. The clip is set to Bruce Springsteen’s 1984 anthem “Born in the U.S.A.” For its half-minute duration, Meta’s multibillionaire chief executive shows off his surf technique.

“Amazing!” commented Lauren Sánchez, the fiancée of Jeff Bezos.

A gaming influencer, @StoneMountain64, wrote, “Now that’s content.”

Mr. Zuckerberg replied, “Just doing my part.”

To Zuck-ologists, the clip was yet another example of the 40-year-old executive’s attempt to remake his image. In recent years, he has gone from a flip-flop-and-hoodie-wearing tech entrepreneur to a sleeker, Richard Bransonesque figure, one who wears Brunello Cucinelli T-shirts, a silver chain and has immersed himself in mixed martial arts.

As one commenter on X put it, “The PR team rehabbing Zuck continues their undefeated streak.”

The video was a sequel of sorts to a video Mr. Zuckerberg posted on July 4, 2021. That one showed him aboard a moving hydrofoil while carrying an American flag to the soundtrack of John Denver’s 1971 hit “Take Me Home, Country Roads.”

The next year he posted a picture of himself wearing an American flag cowboy hat as he grilled sausages. “Smoking these meats,” he wrote in a caption. “Happy 4th!” Last year’s post featured a candid shot of Mr. Zuckerberg and his family.

If social media experts help Mr. Zuckerberg craft his posts, then not much is known about them. Meta representatives have suggested that he does not depend on image consultants. A representative for Meta did not immediately respond to a request to comment.

If the intent behind Mr. Zuckerberg’s patriotic content drops has been to render him more relatable to the American public, despite his approximately $181 billion net worth, according to Bloomberg, they appear to have helped. The online response to this year’s Fourth of July post was largely upbeat, markedly different from the satirical memes that roasted his 2021 hydrofoil post.

But some online observers pointed out an off-note in the clip: its use of “Born in the U.S.A.” Often misinterpreted as a rah-rah anthem, the song tells the story of a Vietnam War veteran who returns home to a lonely welcome and dire circumstances.

Dana White, the chief executive of the Ultimate Fighting Championship, who has long tried to set up a cage match between Mr. Zuckerberg and his tech rival Elon Musk, reacted to the video with a positive comment: “’MERICA!!!!!”

Mr. Musk had a different take, writing on X: “May he continue to have fun on his yachts. I prefer to work.”





Source link

Continue Reading
Advertisement

Trending

Copyright © 2024 World Daily Info. Powered by Columba Ventures Co. Ltd.