Latest news with #TheoVon
CNET
3 hours ago
- CNET
Even The Guy Who Makes ChatGPT Says You Probably Shouldn't Use Chatbots as Therapists
Maybe don't tell your deepest, darkest secrets to an AI chatbot like ChatGPT. You don't have to take my word for it. Take it from the guy behind the most popular generative AI model on the market. Sam Altman, the CEO of ChatGPT maker OpenAI, raised the issue this week in an interview with host Theo Von on the This Past Weekend podcast. He suggested that your conversations with AI should have similar protections as those you have with your doctor or lawyer. At one point, Von said one reason he was hesitant to use some AI tools is because he "didn't know who's going to have" his personal information. "I think that makes sense," Altman said, "to really want the privacy clarity before you use it a lot, the legal clarity." More and more AI users are treating chatbots like their therapists, doctors or lawyers, and that's created a serious privacy problem for them. There are no confidentiality rules and the actual mechanics of what happens to those conversations are startlingly unclear. Of course, there are other problems with using AI as a therapist or confidant, like how bots can give terrible advice or how they can reinforce stereotypes or stigma. (My colleague Nelson Aguilar has compiled a list of the 11 things you should never do with ChatGPT and why.) Altman's clearly aware of the issues here, and seems at least a bit troubled by it. "People use it, young people especially, use it as a therapist, a life coach, I'm having these relationship problems, what should I do?" he said. "Right now, if you talk to a therapist or a lawyer or a doctor about those problems, there's legal privilege for it." The question came up during a part of the conversation about whether there should be more rules or regulations around AI. Rules that stifle AI companies and the tech's development are unlikely to gain favor in Washington these days, as President Donald Trump's AI Action Plan released this week expressed a desire to regulate this technology less, not more. But rules to protect them might find favor. Read more: AI Essentials: 29 Ways You Can Make Gen AI Work for You, According to Our Experts Altman seemed most worried about a lack of legal protections for companies like his to keep them from being forced to turn over private conversations in lawsuits. OpenAI has objected to requests to retain user conversations during a lawsuit with the New York Times over copyright infringement and intellectual property issues. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) "If you go talk to ChatGPT about the most sensitive stuff and then there's a lawsuit or whatever, we could be required to produce that," Altman said. "I think that's very screwed up. I think we should have the same concept of privacy for your conversations with AI that you do with your therapist or whatever." Watch this: OpenAI Debuts "Study Mode" for Students, the Tea App Data Breach, and Could a Robot Dog Deliver Your Next Pizza? | Tech Today 03:26 Be careful what you tell AI about yourself For you, the issue isn't so much that OpenAI might have to turn your conversations over in a lawsuit. It's a question of whom you trust with your secrets. William Agnew, a researcher at Carnegie Mellon University who was part of a team that evaluated chatbots on their performance dealing with therapy-like questions, told me recently that privacy is a paramount issue when confiding in AI tools. The uncertainty around how models work -- and how your conversations are kept from appearing in other people's chats -- is reason enough to be hesitant. "Even if these companies are trying to be careful with your data, these models are well known to regurgitate information," Agnew said. If ChatGPT or another tool regurgitates information from your therapy session or from medical questions you asked, that could appear if your insurance company or someone else with an interest in your personal life asks the same tool about you. "People should really think about privacy more and just know that almost everything they tell these chatbots is not private," Agnew said. "It will be used in all sorts of ways."
The Hindu
a day ago
- Business
- The Hindu
OpenAI CEO Sam Altman flags lack of AI-user privacy; Deepfakes threaten political security; Nvidia orders H20 chips from TSMC
OpenAI CEO Sam Altman flags lack of AI-user privacy Users who treat ChatGPT as a life coach or therapist, sharing extremely personal thoughts or sensitive questions with the AI chatbot, do not have real privacy protections, according to OpenAI CEO Sam Altman. The technologist warned during an episode of 'This Past Weekend' podcast with Theo Von that the AI company could be forced to hand over such content as evidence in case of a lawsuit or a legal proceeding. Altman made it clear that he disagreed with this standard and highlighted the need for a user-AI privacy standard similar to the one that exists between a patient and a doctor or a client and a therapist. Such interactions are generally protected by what is known as 'privilege,' meaning it is hard to use them against the individual in court. This is not merely a hypothetical situation for the company and its users worldwide. OpenAI was critical of The New York Times, claiming that as part of a lawsuit against the AI startup, the media company asked the court to force OpenAI to 'retain all user content indefinitely going forward...' Deepfakes threaten political security Employees who work in U.S. President Donald Trump's administration, high-profile business companies in America, or corporate settings are preparing to tackle the surge in realistic deepfakes across media formats that impersonate their leaders and chiefs. This has been done to U.S. Secretary of State Marco Rubio as well as Trump's chief of staff, Susie Wiles. There is a serious risk that lower-level employees targeted by such scams might disclose highly sensitive information, security details, or trade secrets to malicious attackers representing hostile regimes. Deepfakes can also be used to scam users out of their savings, get access to their passwords, or prevent them from exercising their right to vote unimpeded. However, there are tech experts who believe that the solution to identifying AI-generated deepfakes is deploying more AI tools that can process the details of a deepfake that human viewers might miss out on. Efforts are also on to better track North Korean agents who are embedded in the U.S. workforce. Nvidia orders H20 chips from TSMC Sources have claimed that Nvidia ordered 300,000 H20 chipsets from TSMC last week, with the development being related to a strong Chinese demand for the chips that are used in AI development. H20 was developed by Nvidia for the Chinese market and has less computing power than Nvidia's H100 and its new Blackwell series; it was made after U.S. export restrictions hit its other AI chipsets. Nvidia and TSMC have not yet formally commented on the reports. While U.S. legislators are concerned that China's access to the H20 offerings will affect competition between the two countries when it comes to leading AI technology, Nvidia's perspective is that Chinese interest in its chips will prevent developers from becoming reliant on products from rivals such as Huawei.
Time of India
2 days ago
- Business
- Time of India
'AI will always be smarter than they are': Why Sam Altman thinks college won't matter for his son
CEO doesn't believe his son will go to college for higher education. In a strong and informed opinion about the future of learning, Altman suggested that traditional education, as it's known today, will likely cease to exist in the face of the rapidly advancing age of AI. Tired of too many ads? go ad free now Speaking with comedian Theo Von on This Past Weekend podcast, Altman expressed his personal uncertainty about whether his own child would pursue a college education, emphasising the profound changes AI will bring to both the education system and the employment landscape. Reflecting on the rapid pace of change, he said, 'The last few months have felt very fast. It feels faster and faster, but the last few months have felt very fast. ' This speaks to the accelerating speed at which AI is reshaping industries, including education. Here's the YouTube link to the podcast: Sam Altman | This Past Weekend w/ Theo Von #599 Altman predicts future for higher education When asked whether he anticipated his son attending university, Altman's response was apparent: "Probably not." This distinct assessment comes from someone who left Stanford University in 2005 to pursue entrepreneurial ventures, ultimately building a technology empire. His prediction reflects a broader conviction that AI's expanding influence will fundamentally diminish the necessity and relevance of traditional academic institutions. Altman's scepticism is not just founded on personal considerations but to encompass systemic changes in how society approaches knowledge acquisition and skill development. He debates that the conventional university model already struggles to meet contemporary student needs and will become increasingly obsolete as AI capabilities advance. 18 years from now, the world might be unrecognisable According to Altman, AI integration into daily life will fundamentally modify learning methodologies and how students acquire knowledge. Tired of too many ads? go ad free now "In that world, education is going to feel very different," he said, projecting that by the time his child reaches university age, approximately 18 years from now, educational systems will bear little resemblance to current models. This transformation stems from AI's capacity to process and recall information far beyond human capabilities. Altman contends that students born today will be residents of a world where artificial intelligence consistently outperforms human cognitive abilities, particularly in information processing and retention. In such an environment, he observes, traditional educational objectives centred on intellectual competition and knowledge accumulation will lose their fundamental value. 'We just don't know. We think it's going to be... there are clearly real risks. It kind of feels like you should be able to say something more than that. But in truth, I think all we know right now is that we have discovered, invented, whatever you want to call it, something extraordinary that is going to reshape the course of human history,' Altman noted, reflecting on the profound impact AI will have, not just on education but on society as a whole. AI will always be smarter than students A central element of Altman's argument concerns AI's permanent intellectual advantage over human students. "Kids born today will grow up in a world where AI will always be smarter than they are," he stated. With artificial intelligence embedded across products and services, the traditional educational goal of achieving intellectual superiority through academic achievement becomes meaningless. This cognitive disparity suggests that memorising facts, mastering subjects through rote learning, and competing academically will become increasingly irrelevant pursuits. Instead, Altman implies that educational systems must reimagine their fundamental purposes and methodologies to remain viable in an AI-dominated landscape. AI will enhance not replace Despite his predictions about educational transformation, Altman does not advocate for eliminating human learning entirely. He draws parallels between AI adoption and the introduction of calculators, which transformed mathematical education without rendering it obsolete. "I'm sure the same thing happened with the calculator before, and now this is just a new tool that exists in the tool chain," he explained. This perspective positions artificial intelligence as an enhancement tool that will modify human work and creativity approaches rather than completely replacing human endeavour. Altman suggests that AI will become integrated into educational and professional processes much as calculators became standard mathematical tools, changing methodologies whilst preserving human involvement. Adoption across generations is worrying Altman acknowledges significant generational differences in adapting to AI-driven societal changes. He expresses greater concern for older generations than for young people who will mature alongside these technologies. "I actually think the kids will be fine; I'm worried about the parents," he remarked. This observation reflects historical patterns where younger generations demonstrate superior adaptability to technological innovations whilst older individuals face greater adjustment challenges. Altman's assessment suggests that whilst children will naturally integrate AI into their worldview, adults may struggle with the transition from traditional educational and professional frameworks. Altman is optimistic that human contributions will never be phased out Despite acknowledging AI's cognitive advantages, Altman maintains optimism about distinctly human contributions to future society. He believes that creativity, determination, and social skills will remain essential qualities that artificial intelligence cannot replicate. These characteristics will continue driving progress in ways that transcend technological capabilities. Looking towards long-term societal evolution, Altman suggests that future generations may view our current era as comparatively straightforward. "Looking forward 100 years from now, we may well think the same thing about the future that people in the past thought about us," he noted, indicating that technological adaptation remains a consistent historical pattern.
The Star
4 days ago
- Entertainment
- The Star
Who is watching all these podcasts?
The following are the run times of some recent episodes of several of YouTube's more popular podcasts: 'This Past Weekend w/ Theo Von,' No. 595: Two hours, 14 minutes. 'Club Shay Shay,' No. 172: Two hours, 59 minutes. 'The Shawn Ryan Show,' No. 215: Five hours, four minutes. 'Lex Fridman Podcast,' No. 461: Five hours, 20 minutes. These shows follow the same general format: people sitting in chairs, in generically designed studios, talking. And, like many of the biggest podcasts these days, these shows are all released as videos. They don't feature particularly fancy camerawork, or flashy graphics, or narratives. All of them require time commitments typical of feature films, ballgames or marathon performance art installations. Yet going by YouTube's statistics, hundreds of thousands of people have viewed all of the above episodes. Which leads to comments such as this, as one fan wrote after a recent episode of Von's show: 'Truly, this podcast was amazing to watch.' So a genre of media named for an audio device – the iPod, discontinued by Apple in 2022 – and popularised by audiences enamoured of on-demand listening has transformed in recent years into a visual one. It's well established that the American brain is the prize in a war for attention online, a place that incentivises brief and sensational content, not static five-hour discussions about artificial intelligence. So what gives? Who exactly is watching the supersize video talk shows that have come to define podcasting over the past several years? At the highest level, the audience for video podcasts is simply people who consume podcasts. 'Who is watching these?' said Eric Nuzum, a podcast strategist. 'A person who loves podcasts who happens to be near a screen.' Indeed, according to an April survey by Cumulus Media and media research firm Signal Hill Insights, nearly three-fourths of podcast consumers play podcast videos, even if they minimize them, compared with about one-fourth who listen only to the audio. Paul Riismandel, president of Signal Hill, said this split holds across age groups – it's not simply driven by Generation Z and that younger generation's supposed great appetite for video. But dive a bit deeper into the data, and it becomes clear that how people are watching podcasts – and what counts as watching – is a far more revealing question. According to the Signal Hill survey, about 30% of people who consume podcasts 'play the video in the background or minimise on their device while listening.' Perhaps this person is folding laundry and half-watching 'Pod Save America,' or has 'The Joe Rogan Experience' open in a browser tab while they do busy work at the office. That describes Zoë McDermott, a 31-year-old title insurance producer from Pennsylvania, who said she streams video of Von's show on her phone while she works. 'I don't have the ability to watch the entire thing through, but I do my glance-downs if I hear something funny,' McDermott said. 'It's passive a little bit.' Still, this leaves everyone else – more than half of YouTube podcast consumers, who say they are actively watching videos. Here, it gets even trickier. YouTube, the most popular platform for podcasts, defines 'views' in a variety of ways, among them a user who clicks 'play' on a video and watches for at least 30 seconds: far from five hours. And the April survey data did not distinguish between people who were watching, say, four hours of Lex Fridman interviewing Marc Andreessen from people who were viewing the much shorter clips of these podcasts that are ubiquitous on TikTok, Instagram Reels, X and YouTube itself. All of which makes it hard to pinpoint a 'typical' podcast viewer. Is it a couple on the couch with a bucket of popcorn, streaming to their smart TV? Is it a young office worker scrolling through TikTok during his commute? Or is it the same person engaging in different behaviour at different points in the day? Alyssa Keller, who lives in Michigan with her family, said sometimes she watches 'The Shawn Ryan Show' on the television with her husband. But more often, she puts the video on the phone for a few hours while her children are napping. This means she sometimes has to watch marathon episodes in chunks. 'I've been known to take multiple days,' she said. 'Nap times only last for like two hours.' In February, YouTube announced that more than 1 billion people a month were viewing podcasts on its platform. According to Tim Katz, head of sports and news partnerships at YouTube, that number is so large that it must include users who are actually mainlining five-hour talk shows. 'Any time you have a number that large, you're going to have a broad swath of people consuming in lots of different ways,' Katz said. Recently, The New York Times asked readers if and how they consume video podcasts. Many of the respondents said they played video podcasts in the background while attending to work or chores, and still treated podcasts as audio-only products. A few said they liked being able to see the body language of podcast hosts and their guests. Still others said that they didn't like video podcasts because they found the visual component distracting or unnecessary. Video can have its drawbacks. Lauren Golds, a 37-year-old researcher based in Virginia, said she regularly hate-watches podcasts at work – in particular 'On Purpose,' which is hosted by British entrepreneur and life coach Jay Shetty. She said she had had awkward encounters when co-workers have looked at her screen and told her that they love the show she's watching. 'There's no way to say it's garbage and I'm watching it for entertainment purposes to fill my need for hatred,' Golds said. One thing a 'typical' podcaster consumer is less likely to be these days is someone listening to a full-attention-required narrative program. Say 'podcast' and many people still instinctively think of painstakingly produced, deeply reported, audio-only shows such as 'Serial' and 'This American Life,' which listeners consumed via audio-only platforms such as Apple Podcasts and the iHeartRadio app. Traditional podcasts relied on host-read and scripted ads to make money, and on media coverage and word of mouth for discovery. And it was a lot of money, in some cases: In 2019, to take one example, Spotify acquired Gimlet – one of the defining podcast producers of the 2010s – as part of a US$340mil (RM1.4bil) investment in podcast startups. Now, the size of the market for video podcasts is too large to ignore, and many ad deals require podcasters to have a video component. The platforms where these video podcasts live, predominantly YouTube and Spotify, are creating new kinds of podcast consumers, who expect video. McDermott, the Von fan, said the video component made her feel like she had a friendly guest in her home. 'It feels a little more personal, like somebody is there with you,' she said. 'I live alone with my two cats and I'm kind of in a rural area in Pennsylvania, so it's just a little bit of company almost.' The world of podcasts today is also far more integrated into social media. Clips of video podcasts slot neatly into the Gen Z and millennial behemoths of TikTok and Instagram. The sophisticated YouTube recommendation algorithm suggests relevant new podcasts to viewers, something that wasn't possible in the old, siloed model on other platforms. To get a sense of just how much things have changed, imagine the viral podcast appearances of the 2024 presidential campaign – Donald Trump on Von's podcast and Kamala Harris on 'Call Her Daddy' – happening without YouTube, TikTok, Instagram and X. You can't. In a sign of the times, in June, radio company Audacy shuttered Pineapple Street Studios, a venerable podcast producer known for its in-depth narrative shows such as 'Wind of Change' and Ronan Farrow's 'The Catch and Kill Podcast.' Jenna Weiss-Berman, who co-founded Pineapple Street, is now head of audio at comedian and actress Amy Poehler's Paper Kite Productions. Poehler's new podcast, 'Good Hang with Amy Poehler,' is typical of the genre: a charismatic, well-known host, interviewing other charismatic, well-known people. Weiss-Berman said she was concerned that the costs associated with high-quality video production would be prohibitive for smaller podcast creators, who faced almost no barrier to entry when all the genre required was a few microphones. 'If you want to do it well, you need a crew and a studio,' Weiss-Berman said. For podcasters with an established audience, the potential of video to open up new audiences for the world of talk podcasts is obvious. (The Times has introduced video podcasts hosted by some of its more recognisable columnists.) Adam Friedland, a comedian who started his video interview show in 2022, first came to prominence on an irreverent and lewd audio-only hangout podcast with two fellow comedians. He got an early taste of the limitations of traditional podcast distribution when he discovered fan cutups of the funniest moments of his old show on YouTube. 'There was an organic growth to it,' Friedland said. 'We weren't doing press or promoting it.' Friedland's new show is an arch interview program with high-profile guests and considerably fewer impenetrable – not to mention scatological – references. Along with that, distribution over YouTube has made a once cult figure something a bit closer to a household name, as he discovered recently. 'There was a regular middle-aged guy at a Starbucks who said he liked the show,' Friedland recalled. 'Some guy holding a Sweetgreen.' Friedland's show is the rare video podcast with a distinctive visual point of view. The vintage-looking set is a reconstruction of 'The Dick Cavett Show.' And Friedland made it clear that he prefers people to watch the show rather than listen to it. The many ways that Americans now consume podcasts – actively and passively, sometimes with another device in hand, sometimes without – bears an obvious similarity to the way Americans consume television. 'I think podcasts could become kind of the new basic cable television,' said Marshall Lewy, chief content officer of Wondery, a podcast network owned by Amazon. Think: shows that are cheaper to produce than so-called premium streaming content, consumed by audiences used to half-watching television while scrolling their smartphones, in a wide variety of genres. Indeed, although talk dominates among video podcasts, Lewy said he thought the trend for video would lead to more shows about food and travel – categories beloved by advertisers – that weren't ideal when podcasts were audio only. All of which calls into question the basic nature of the term 'podcast.' Riismandel, who runs the research firm Signal Hill, said he thought the category applied to any programming that could be listened to without video and still understood. According to Katz, the YouTube executive, the nature of the podcaster is undergoing a redefinition. It includes both audio-only podcasters moving to video, as well as social media content creators who have realised that podcasts present another opportunity to build their audiences. One concern with the shift to video, according to former Vox and Semafor video boss Joe Posner, is that people who are less comfortable on screen will be left out. This could lead to a deepening gender divide, for example, since women are much more likely to face harassment over their looks, especially from an engaged online fan base – and therefore potentially less likely to want to be on camera for hours on end. Still, for all the eyeballs moving to YouTube, audio remains the way most consumers experience podcasts, according to the April survey, with 58% of people listening to only audio or to a minimised or backgrounded video. And although YouTube is now the most used platform for podcast consumption, per the survey, it's far from monolithic; a majority of podcast consumers say they use a platform other than YouTube most often, whether it's Spotify or Apple Podcasts. That's why at least one pillar of audio-first podcasting doesn't see much to be alarmed about. Ira Glass, creator of the foundational long-form radio show 'This American Life,' said the fact that the podcast tent has gotten bigger and thrown up a projector screen doesn't threaten a program like his. 'That's a strength, not a weakness – that both things exist and are both called the same thing,' Glass said. He stressed that audio-only podcasting has formal strengths that video podcasts don't. 'There's a power to not seeing people,' Glass said. 'There's a power to just hearing things. It just gets to you in a different way. But if people want to watch people on a talk show, that seems fine to me. I don't feel protective of podcasting in that way. I don't have snowflake-y feelings about podcasts.' – ©2025 The New York Times Company This article originally appeared in The New York Times.
CNET
5 days ago
- CNET
Even OpenAI's CEO Says Be Careful What You Share With ChatGPT
Maybe don't spill your deepest, darkest secrets with an AI chatbot. You don't have to take my word for it. Take it from the guy behind the most popular generative AI model on the market. Sam Altman, the CEO of ChatGPT maker OpenAI, raised the issue this week in an interview with host Theo Von on the This Past Weekend podcast. He suggested that your conversations with AI should have similar protections as those you have with your doctor or lawyer. At one point, Von said one reason he was hesitant to use some AI tools is because he "didn't know who's going to have" his personal information. "I think that makes sense," Altman said, "to really want the privacy clarity before you use it a lot, the legal clarity." More and more AI users are treating chatbots like their therapists, doctors or lawyers, and that's created a serious privacy problem for them. There are no confidentiality rules and the actual mechanics of what happens to those conversations are startlingly unclear. Of course, there are other problems with using AI as a therapist or confidant, like how bots can give terrible advice or how they can reinforce stereotypes or stigma. (My colleague Nelson Aguilar has compiled a list of the 11 things you should never do with ChatGPT and why.) Altman's clearly aware of the issues here, and seems at least a bit troubled by it. "People use it, young people especially, use it as a therapist, a life coach, I'm having these relationship problems, what should I do?" he said. "Right now, if you talk to a therapist or a lawyer or a doctor about those problems, there's legal privilege for it." The question came up during a part of the conversation about whether there should be more rules or regulations around AI. Rules that stifle AI companies and the tech's development are unlikely to gain favor in Washington these days, as President Donald Trump's AI Action Plan released this week expressed a desire to regulate this technology less, not more. But rules to protect them might find favor. Read more: AI Essentials: 29 Ways You Can Make Gen AI Work for You, According to Our Experts Altman seemed most worried about a lack of legal protections for companies like his to keep them from being forced to turn over private conversations in lawsuits. OpenAI has objected to requests to retain user conversations during a lawsuit with the New York Times over copyright infringement and intellectual property issues. (Disclosure: Ziff Davis, CNET's parent company, in April filed a lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems.) "If you go talk to ChatGPT about the most sensitive stuff and then there's a lawsuit or whatever, we could be required to produce that," Altman said. "I think that's very screwed up. I think we should have the same concept of privacy for your conversations with AI that you do with your therapist or whatever." Be careful what you tell AI about yourself For you, the issue isn't so much that OpenAI might have to turn your conversations over in a lawsuit. It's a question of whom you trust with your secrets. William Agnew, a researcher at Carnegie Mellon University who was part of a team that evaluated chatbots on their performance dealing with therapy-like questions, told me recently that privacy is a paramount issue when confiding in AI tools. The uncertainty around how models work -- and how your conversations are kept from appearing in other people's chats -- is reason enough to be hesitant. "Even if these companies are trying to be careful with your data, these models are well known to regurgitate information," Agnew said. If ChatGPT or another tool regurgitates information from your therapy session or from medical questions you asked, that could appear if your insurance company or someone else with an interest in your personal life asks the same tool about you. "People should really think about privacy more and just know that almost everything they tell these chatbots is not private," Agnew said. "It will be used in all sorts of ways."
