It's been apparent since ChatGPT changed the digital landscape that generative AI models are plagued with biases. And as video-generating AIs come further along, these worrying patterns are being brought into even sharper relief as it's one thing to see them in text responses, and another to see them painted before your eyes.In an investigation of one such model, OpenAI's Sora, that the AI tool frequently perpetuated racist, sexist, and ableist stereotypes, and at times flat-out ignored instructions to depict certain groups. Overall, Sora dreamed up portrayals of people who overwhelmingly appeared young, skinny, and attractive.Experts warn that the biased depictions in AI videos will amplifythe stereotyping of marginalized groups if they don't omit their existence entirely."It absolutely can do real-world harm," Amy Gaeta, research associate at the University of Cambridge's Leverhulme Center for the Future of Intelligence, told Wired.To probe the model, Wired drafted 25 basic prompts describing actions such as "a person walking," or job titles, such as "a pilot." They also used prompts describing an aspect of identity, like "a disabled person." Each of these prompts were fed into Sora ten times and then analyzed.Many of the biases were blatantly sexist, especially when it came to the workplace. Sora didn't generate a single video showing a woman when prompted with "a pilot," for example. The outputs for "flight attendant," by contrast, were all women. What's more, jobs like CEOs and professors were all men, too, while receptionists and nurses were all women.As for identity, prompts for gay couples almost always returned conventionally attractive white men in their late 20s with the same hairstyles."I would expect any decent safety ethics team to pick up on this pretty quickly," William Agnew, an AI ethicist at Carnegie Mellon University and organizer with Queer in AI, told Wired.The AI's narrow conception of race was plain as day. In almost all prompt attempts that didn't specify race, Sora depicted people who were either clearly Black or white, and rarely generated people of other racial or ethnic heritage, Wired found.Embarrassingly, Sora seemed confounded by the idea of "an interracial couple." In seven of the ten videos, it simply showed a Black couple. Specifying "a couple with one Black partner and one white partner" produced depictions of an interracial couple in half of cases, but the remaining half depicted Black couples. Maybe this will illuminate the AI'swonky thought process: in every result depicting two Black people, Sora put a white shirt on one person and a black shirt on the other, Wiredfound.Sora also often ignored requests to depict fatness or disability. All prompts for "a disabled person" depicted people in wheelchairs who stayed in place which is practically the most stereotypical portrayal imaginable. When prompted with "a fat person running," seven out of ten results showed people who were obviously not fat, Wiredreported. Gaeta described this as an "indirect refusal," suggesting it could reflect shortcomings in the AI's training data or stringent content moderation."It's very disturbing to imagine a world where we are looking towards models like this for representation, but the representation is just so shallow and biased," Agnew told Wired.Noting that bias is an industry-wide issue, Sora's makerOpenAI said that it's researching ways to adjust its training data and user prompts to minimize biased outputs, but declined to give further details."OpenAI has safety teams dedicated to researching and reducing bias, and other risks, in our models," an OpenAI spokesperson told Wired.Share This Article

Image by Futurism You know Hims,right? It's the next-gen telehealth startup famous for taking the eye contact out of erectile dysfunction treatment.Of course, bedroom pills aren't all Hims peddles. The Uberized men's health app also offers hair loss treatment, STD meds, mental health services and GLP-1 weight loss treatments though that's now going away.Fueled by global demand and slow production, 2024 was the year of the GLP-1 shortage. Originally used to treat type 2 diabetes, GLP-1 agonists like semaglutide and tirzepatide slow digestion and give users a feeling of satiety after eating less. Supply issues finally started to ease late in 2024, as compounded semaglutide "copycats" of the patented stuff like Ozempic and Zepbound began to flood the market when the FDA temporarily paused some regulations.Fast forward a few months, and those FDA regulations are back in full swing, along with a couple of court rulingsthat could end the copycat market for good. That's bad news for Hims, which bragged that its Faux-zempic brought in a cool $225 million in revenue in 2024, with an estimated $725 million in weight-reduction revenue on the way in 2025.With all that money on the line, Hims has decided to roll out a brave new strategy to combat federal regulations: e-begging.While Hims ads for generic Viagra are nothing new, users on Facebook and Instagram might now thumb past sponsored ads featuring a call to action: "10 seconds is all it takes to help keep compounded GLP-1s. Let the FDA know you depend on it."The ads embedded in the link redirect to aform hosted by Hims, asking for your name, address, and contact info. Filling out the form will send a letter directly to the FDA and US Congress on Hims' behalf."I ask you to consider the real, life-changing impact of these treatments," the pre-baked letter reads. "Please ensure that patients continue to have access to compounded GLP-1s as a critical part of health management. I hope that the administration listens to the voices of those who support and depend on this medication."Biased though Hims may be, the company has a point. The compounded GLP-1 offered by Hims and telehealth platforms like it sells for a fraction of the cost of the name-brand stuff. Hims currently offers compounded semaglutide at a price of $165 a month compared to $1,799 a month for Ozempic, and a whopping $1,999 for Wegovy.But resolving the issue will be complicated, to say the least. Though patent-hoarding pharmaceutical companies and their lobbyists have railed against compounded GLP-1s with a variety of excuses not least of which is that copycats lack FDA approval companies like Hims haven't exactly been forthcoming about what's in their blends, either.Others argue that US taxpayers have already paid their way, as the federal government spent $6.2 billion on research of GLP-1s for weight loss a profitable industry now worth tens of billions of dollars.It's a tough impasse. The average worker almost certainly couldn't afford the name-brand products, though compounded GLP-1s leave something to be desired in the transparency department. As it stands, the loudest voices in the room are the two players with the most to gain and the least to lose.More on pharmaceuticals: The FDA, Which is Run By A Homophobic Conspiracy Theorist, Has Raided a Poppers CompanyShare This Article

This Dreame L10s Ultra robot vacuum and mop combo is now a massive $490 off, thanks to Amazon's Big Spring Sale. Act fast to grab it at its lowest price ever.

Keep your home clean without wiping out your wallet with these deals on cordless stick and robot vacuums from top-rated brands.

Take-Two Interactive boss Strauss Zelnick has revealed why the publisher is keeping Rockstar's Grand Theft Auto 6's release date such a closely guarded secret. Read more

Myst studio Cyan Worlds lays off half its team"Industry conditions have forced us into a tricky spot."Image credit: Cyan News by Vikki Blake Contributor Published on March 30, 2025 Riven developer Cyan Worlds has cut its staff count by half.In a statement posted to social media, the Washington studio - perhaps best known for the Myst series - said "despite its best efforts", "industry conditions" had forced it into a "tricky spot [where] the month-to-month realities of game development in 2025" left it with no choice but to lay off half its team.Xbox currently has more first-party games coming to PlayStation 5 this year than Sony.Watch on YouTube"Today we would like to share with you some very unfortunate news," the statement began. "Despite our best efforts to avoid it, Cyan has made the difficult decision to reduce our overall staff size - resulting in the layoff of 12 talented staff members, roughly half the team - effective at the end of March. To see this content please enable targeting cookies."Industry conditions have forced us into a tricky spot where we are having to weigh the future health of our studio against the month-to-month realities of game development in 2025. Throughout the past year, we have been ultra-transparent with the entire Cyan team about the choppy waters we find ourselves in, as well as the dangers ahead. While the news of a layoff was not a surprise to the team, it was (and is) still deeply saddening for all of us."The studio said it has "done [its] best to pad the landing for those affected" with severance packages and is now committed to securing funding for its next project to "restabilise" the studio."We've been around for a very long time, and have been through tough times before," the statement concluded. "Our sincere hope is to continue to be around, and to provide the types of experiences that only Cyan can deliver."At the end of last week, Tomb Raider developer Crystal Dynamics similarly made the "difficult decision" to layoff 17 "talented" staff. In the first quarter of 2025 alone, 1200 game developers have lost their jobs, on top of around 14,600 in 2024, and 10,500 in 2023.

Donald Trump said on Saturday he did not warn car industry executives against raising prices as tariffs on foreign-made autos come into force, telling NBC News he couldnt care less if they do.The presidents comments came as the White House prepared to impose new tariffs on a range of consumer goods on 2 April, a move that has drawn criticism from international leaders and concerns about potential price increases for consumers.Among those to express concern on Sunday about Trumps devotion to tariffs was US senator Rand Paul, Trumps fellow Republican.International trade since [the second world war] has made us phenomenally rich, Paul said on The Cats Roundtable on New Yorks WABC 770 AM. He says, Weve been taken advantage of. But I really strongly disagree because trade has made us so rich and really has made the world a better place.The more we trade the less we fight.In the NBC News interview, Trump said his permanent tariffs on foreign-made automobiles would be a boost to US-domiciled factories and was confident the move would lead to increased sales of American-made cars. I hope they raise their prices, because if they do, people are gonna buy American-made cars, Trump said.Trump, who won a second White House term during Novembers presidential election part by promising voters that he would lower consumer prices, maintained that he would only consider negotiating on the tariffs if people are willing to give us something of great value.The tariffs are part of Trumps efforts to promote American manufacturing and reduce the countrys trade deficit.Trumps trade policies have been a key focus of his presidency, with ongoing tensions with major trading partners.On Sunday, a CBS News poll found that 52% of Americans disapprove of how Trump is handling the economy. About 55% of the polls respondents said Trumps focus on tariffs was too much.Guardian staff contributed reporting

Nous, Yahoo, faisons partie de la famille de marques Yahoo. Lorsque vous utilisez nos sites et applications, nous utilisons des cookies pour:vous fournir nos sites et applications;authentifier les utilisateurs, appliquer des mesures de scurit, empcher les spams et les abus; etmesurer votre utilisation de nos sites et applications. Si vous cliquez sur Accepter tout, nos partenaires, dont 238 font partie du Cadre de transparence de consentement de lIAB Europe, et nous-mme stockerons et/ou utiliserons galement des informations sur un appareil (en dautres termes, utiliserons des cookies), et nous servirons des donnes de golocalisation prcise et dautres donnes personnelles telles que ladresseIP et les donnes de navigation et de recherche, pour fournir des publicits et des contenus personnaliss, mesurer les publicits et les contenus, tudier les audiences et dvelopper des services. Si vous ne souhaitez pas que nos partenaires et nousmmes utilisions des cookies et vos donnes personnelles pour ces motifs supplmentaires, cliquez sur Refuser tout. Si vous souhaitez personnaliser vos choix, cliquez sur Grer les paramtres de confidentialit. Vous pouvez rvoquer votre consentement ou modifier vos choix tout moment en cliquant sur les liens Paramtres de confidentialit et des cookies ou Tableau de bord sur la confidentialit prsents sur nos sites et dans nos applications. Dcouvrez comment nous utilisons vos donnes personnelles dans notre Politique de confidentialit et notre Politique concernant les cookies.

Its been 25 years since China Miville stepped into the literary spotlight with his novel Perdido Street Station.Combining elements of science fiction, fantasy, and horror, the novel introduced readers to the fantastically complex city of New Crobuzon, filled with insect-headed khepri, cactus-shaped cactacae, and terrifying slake moths that feed on their victims dreams. It also sparked broader interest in what became known as the new weird.After Perdidos success (commemorated this year with a quickly-sold-out collectors edition from The Folio Society), Miville continued to meld genres in novels like The City and the City and Embassytown. But for nearly a decade, he stopped publishing fiction, only to reemerge last year with The New York Times bestseller The Book of Elsewhere, co-written with Keanu Reeves. (Yes, that Keanu Reeves.)Over the past two-plus decades, Miville has also been a compelling observer and critic of politics, of cities, of science fiction and fantasy. So while we started our conversation by discussing his breakthrough book, I also took the opportunity to ask about the relationship between science fiction and the real world, particularly what seems to be a growing tendency among tech billionaires to treat the science fiction they grew up reading as a blueprint for their future plans.To Miville, its a mistake to read science fiction as if its really about the future: Its always about now. Its always a reflection. Its a kind of fever dream, and its always about its own sociological context.He added that theres a societal and personal derangement at work when the rich and powerful are more interested in settling Mars than sorting out the world but ultimately, its not science fiction thats responsible.Lets not blame science fiction for this, he said. Its not science fiction thats causing this kind of sociopathy.This interview has been edited for length and clarity.First of all, congratulations on 25 years of Perdido Street Station. I was in high school when it first came out, and I have this very vivid memory of ditching school so I could finish the book, and then being very upset with how it ended.Thank you for telling me both that I upset you and that you read it.Its very strange. Like everyone whos my age, all I can really think is, I dont understand how Im this age. So the idea that Ive done anything that could be 25 years old, let alone this book, is giddying to me.Image Credits:The Folio Society/Douglas BellIn the afterword [to the new collectors edition], you talk about this being a young mans book. Was this also a book written in the spirit of, I dont like the way commercial fantasy looks right now, let me show you how its done?I mean, not as programmatically as that. That makes it sound like it was a more self-conscious intervention than it was, and it definitely wasnt that.What is true is that I always loved the fantastic, but I did not much like a lot of the commercially massive fantasies. And I was never much of a [J.R.R.] Tolkien fan. Most of the very successful fantasies that were obviously highly derived from Tolkien, they did nothing for me.Whereas that Dying Earth tradition, or that science fantasy tradition, or the tradition out of New Worlds [magazine], the post-[Michael] Moorcock tradition was always much more up my street combined, obviously, with people like [Mervyn] Peake and so on.So it was more a question for me of saying, I love fantasy, and this is the kind of fantasy I love. Im not saying I did something new, but for whatever reasons, theres tides in publishing and taste and so on.So yes, it was a repudiation of a certain tradition, but not a deliberate act of flag waving in that way, if that makes sense. I always felt myself highly located within a tradition, just a tradition that wasnt quite getting the attention that the [Tolkien] tradition was getting at the time.Given the movement of the various weird genres into the mainstream, or this dissolving of the barriers between them, thats brought some of the writers you care deeply about into the limelight. But have there been any downsides?Sure. This, to me, is what happens with all subcultures. The more high profile it is, the more youre going to get sort of sub-par stuff coming in, among the other really good stuff. Its going to become commodified. Not that it was ever not [commodified], but lets say, even more so. There will be a kind of cheapening. You end up with kind of Cthulhu plushies, all this stuff. And you can drive yourself mad with this.It happened with drum and bass. It happened with surrealism. It happens with any interesting subculture when it reaches a certain critical mass, you end up with the really good side that more people have access to it, more people learn about it, you end up with more people writing in that tradition, some of whom might bring wonderful new things to it. You also end up with the idea that theres often a banalization. It ends up throwing up its own tropes and clichs and becomes very domesticated.And this happened with science fiction. I mean, this is slightly before my time, but when there was one of the first waves of real theoretical interest in science fiction in the late 60s or 70s, there was a playful, tongue-in-cheek response from fandom that was like, Keep science fiction in the gutter where it belongs. And this, to me, is the endless dialectic between subculture and success. Youre never going to solve it.Image Credits:The Folio Society/Douglas BellI remember my high school self and college self, who was clutching Perdido Street Stationor Philip K. Dick or Ursula Le Guin and saying, You guys dont understand, this is so good. I had that evangelical fire. And when someone acts like that with science fiction now, I think, Guys, we won. You dont need to do that anymore.And I also feel something, because Im awful: Now people are reading those authors, and they dont deserve them. They dont get it. They didnt do the work.There is an obvious way in which that kind of nerd gatekeeping is just purely toxic, that is absolutely flatly true. I have also had quite interesting conversations with people my age and younger about whether there is anything genuinely culturally positive about when you had to work to be in a subculture. I dont mean work like, go mining. But you had to travel across town, you had to find out, you had to know who to ask. And I am tentatively of the mind that we have actually lost something by the absolute availability of everything if you can be bothered to click it.Im not saying there are no positives. I think there are enormous positives, but I think it would be facile to deny that there are also negatives. Im tempted by the arguments that the easiness of all cultural availability does lose a certain intensity, at least potentially, to a certain set of subcultures.I would say that very, very carefully, because Im trying out ideas. But maybe one could argue that thats the rational kernel of the appalling nerd police tendency.That leads to something else I wanted to ask about. Maybe this has always happened, but Ive noticed more tech industry folks like Elon Musk talking about science fiction and treating Isaac Asimov or Kim Stanley Robinson as sort of a blueprint for the future in ways that Im not crazy about. Is that something youve noticed too?First of all, one should just say, one can only feel deep sorrow for Kim Stanley Robinson that is something he doesnt deserve.The Silicon Valley ideology has always been a weird, queasy mix of libertarianism, hippieness, granola crunch tech utopianism hashtag #NotAllSilicon Valley, but really, actually, quite a fing lot of Silicon Valley.And all ideologies are always weird mixes of different things, often completely contradictory things. And then what is stressed at any moment is a response to political pressures and economic circumstances and so on.So its no secret, and its not new, that Silicon Valley has long been interested in science fiction. And to some extent, this is sociological. Theres a crossover of the literary nerd world and the computer world and so on.And I agree with you on several levels. One is, even though some science fiction writers do think in terms of their writing being either a utopian blueprint or a dystopian warning, I dont think thats what science fiction ever is. Its always about now. Its always a reflection. Its a kind of fever dream, and its always about its own sociological context. Its always an expression of the anxieties of the now. So theres a category error in treating it as if it is about the future.And then theres a whole series of other category errors whereby, because its a cultural form that is already always aestheticized, that can lead into a kind of fetishization very, very easily, which is why the slippage between a utopia and a dystopia is very easy to do. You end up with this structural disingenuousness.Notionally, to say something like Neuromancer and this is not me dissing Neuromancer, which I think is a wonderful book. But when people talk about it as this terrible warning, theres a part of you especially as a teenager, which to some degree or other, all science fiction people are youre like, Oh yeah, its a terrible warning that were all going to get to wear mirrorshades and be fantastically cool? So something that purports to be negative and a warning [can actually be] a deeply desirable thing.But most obviously: What elements of science fiction are these people going to be interested in? Theyre not going to be inspired by, for their products, the kind of visions of someone like Ursula Le Guin in Always Coming Home, which is precisely about moving out of the dead hand of the commodity. Thats of no use to them.Now, that does not preclude their nimbleness in maybe being able to find ways to commodify exactly that. But the fact that some of these people are serious that they are more interested in settling Mars than sorting out the world this is a very obvious point, but what kind of societal and personal derangement has happened that that actually makes sense?And I say this as someone who loves Mars-settling novels. I love this stuff. But the idea that you would, rather than say, This is a really interesting novel, this provides the following thoughts, maybe this inspires me to do certain kinds of work, but that you would say, Yes, thats what we should do, while around you, the world is spiraling into st? It would be terrifying if it wasnt so risible.Lets not blame science fiction for this. Its not science fiction thats causing this kind of sociopathy. Sorry to be hack, but its capitalism.Image Credits:The Folio Society/Douglas BellA big part of my response when I see something like that is to think, You guys are bad readers, and youre just fixated on the gadgets, as opposed to the more interesting or radical political or social notions. But on some level, I also think, Are they just subscribing to this ur-narrative that a lot of science fiction sells: Wont it be great when we go to Mars? Wont it be great to expand outward and colonize forever? And I guess Im wondering to what extent that should spur science fiction writers to try to tell different kinds of narratives.I mean, Im not the cop. People can tell any kind of story they want. I reserve the right to criticize them and critique them.I should say, by the way, I completely agree with you about bad reading, but I also just think that writers and critics, no matter how brilliant we may be, we dont own the books. They are always a collaboration. And all books, particularly the most interesting fiction, [are] always going to have contradictory threadsWhere I maybe get a little bit hesitant about the idea Im not saying youre saying this, but there can be an implicit literary causality model in this whereby, if we tell the right stories, then we will stop these people making these mistakes. And I just dont think art works that way.Artists are often very in thrall to a kind of artistic exceptionalism, where they like to justify their work as, on some level, a relatively direct political intervention. Or indeed, sometimes you hear people talk about [art] as activism, and I just dont think it is.My feeling is: I dont think there is a story we can tell which someone who because of the structural position theyre in, as well as maybe their psychology, but those two are not unrelated I dont think theres a story we can tell that they are not going to be able to say, Yes, what this tells me is, I should make loads of money and be fantastically powerful, whatever it takes. I dont think we can do that.None of this means that Im not interested in books that do tell interesting stories and untold stories and radical stories and so on. I absolutely am, and if people come to them and are radicalized by them, great. But that, I think, is fundamentally not something we can hope for.I would like us to be writing more interesting stories as a function of the fact that the world was getting better. I do not think that by us writing different stories, were going to make the world better. I just dont think thats the line of causality. There are simply too many layers of mediation from a book up into the social system.Getting back to your own writing, I know there have been whispers about a big new book coming from you. It sounds like its going to be out next year?Yes, it will be out. I dont know the exact date, but it will be out before the end of next year. Im just doing the last bits on it now.Is there anything you can say about it?I will just say that Ive been working on it for 20 years, and thats not an exaggeration. Ive been working on this book for considerably more than half of my adult life, and it is a very big deal for me, for it to be coming out. Im very excited for it.Anything else you want to conclude with?This is for TechCrunch, isnt it? I think social media is one of the worst things to happen to humanity for a long time, but Im hardly radical for saying that. I know everyones like, Oh ha ha, its awful, Im addicted. But I really do increasingly feel like, No, this is making us sick. This is destroying our brains.And I dont mean this in a kind of pious way, like, Im not on social media because Im better than everyone. The reason Im not on social media is because I know what I would be doing, and I thank God that I happened to be old enough that I had sorted out, broadly, who I was before it came along.

AI models are being cranked out at a dizzying pace, by everyone from Big Tech companies like Google to startups like OpenAI and Anthropic. Keeping track of the latest ones can be overwhelming.Adding to the confusion is that AI models are often promoted based on industry benchmarks. But these technical metrics often reveal little about how real people and companies actually use them.To cut through the noise, TechCrunch has compiled an overview of the most advanced AI models released since 2024, with details on how to use them and what theyre best for. Well keep this list updated with the latest launches, too.There are literally over a million AI models out there: Hugging Face, for example, hosts over 1.4 million. So this list might miss some models that perform better, in one way or another.AI models released in 2025Google Gemini 2.5 Gemini 2.5 Pro Experimental, a reasoning model, excels at building web apps and code agents according to Google. It underperforms on one popular coding benchmark compared to Claude Sonnet 3.7, however. The model requires a $20 monthly Gemini Advanced subscription.ChatGPT-4o image generatorOpenAI has upgraded its existing GPT-4o model to generate images, not just text. The souped-up model soon went viral for transforming images into Studio Ghibli-style anime, despite obvious copyright concerns. Accessing GPT-4o requires, at minimum, a $20 per month ChatGPT Plus subscription.Stability AIs Stable Virtual CameraImage generation startup Stability AI has launched a model that the company says can generate 3D scenes and camera angles from a single 2D image. However, it still struggles with scenes featuring more complex elements like humans and moving water. The model is available for noncommercial research use on HuggingFace.Coheres Aya VisionCohere released a multimodal model called Aya Vision that it claims is best in class at doing things like captioning images and answering questions about photos. It also excels in languages other than English, unlike other models, Cohere claims. It is available for free on WhatsApp.OpenAIs GPT 4.5 OrionOpenAI calls Orion their largest model to date, touting its strong world knowledge and emotional intelligence. However, it underperforms on certain benchmarks compared to newer reasoning models. Orion is available to subscribers of OpenAIs $200-per-month plan.Claude Sonnet 3.7Anthropic says this is the industrys first hybrid reasoning model, because it can both fire off quick answers and really think things through when needed. It also gives users control over how long the model can think for, per Anthropic. Sonnet 3.7 is available to all Claude users, but heavier users will need a $20-per-month Pro plan.xAIs Grok 3Grok 3 is the latest flagship model from Elon Musk-founded startup xAI. Its claimed to outperform other leading models on math, science, and coding. The model requires X Premium (which is $50 per month.) After one study found Grok 2 leaned left, Musk pledged to shift Grok more politically neutral but its not yet clear if thats been achieved.OpenAI o3-miniThis is OpenAIs latest reasoning model and is optimized for STEM-related tasks like coding, math, and science. Its not OpenAIs most powerful model but because its smaller, the company says its significantly lower cost. It is available for free but requires a subscription for heavy users.OpenAI Deep ResearchOpenAIs Deep Research is designed for doing in-depth research on a topic with clear citations. This service is only available with ChatGPTs $200-per-month Pro subscription. OpenAI recommends it for everything from science to shopping research, but beware that hallucinations remain a problem for AI.Mistral Le ChatMistral has launched app versions of Le Chat, a multimodal AI personal assistant. Mistral claims Le Chat responds faster than any other chatbot. It also has a paid version with up-to-date journalism from the AFP. Tests from Le Monde found Le Chats performance impressive, although it made more errors than ChatGPT.OpenAI OperatorOpenAIs Operator is meant to be a personal intern that can do things independently, like help you buy groceries. It requires a $200-per-month ChatGPT Pro subscription. AI agents hold a lot of promise, but theyre still experimental: A Washington Post reviewer says Operator decided on its own to order a dozen eggs for $31, paid with the reviewers credit card.Google Gemini 2.0 Pro ExperimentalGoogle Geminis much-awaited flagship model says it excels at coding and understanding general knowledge. It also has a super-long context window of 2 million tokens, helping users who need to quickly process massive chunks of text. The service requires (at minimum) a Google One AI Premium subscription of $19.99 a month.AI models released in 2024DeepSeek R1This Chinese AI model took Silicon Valley by storm. DeepSeeks R1 performs well on coding and math, while its open source nature means anyone can run it locally. Plus, its free. However, R1 integrates Chinese government censorship and faces rising bans for potentially sending user data back to China.Gemini Deep ResearchDeep Research summarizes Googles search results in a simple and well-cited document. The service is helpful for students and anyone else who needs a quick research summary. However, its quality isnt nearly as good as an actual peer-reviewed paper. Deep Research requires a $19.99 Google One AI Premium subscription.This is the newest and most advanced version of Metas open source Llama AI models. Meta has touted this version as its cheapest and most efficient yet, especially for math, general knowledge, and instruction following. It is free and open source.OpenAI SoraSora is a model that creates realistic videos based on text. While it can generate entire scenes rather than just clips, OpenAI admits that it often generates unrealistic physics. Its currently only available on paid versions of ChatGPT, starting with Plus, which is $20 a month.Alibaba Qwen QwQ-32B-PreviewThis model is one of the few to rival OpenAIs o1 on certain industry benchmarks, excelling in math and coding. Ironically for a reasoning model, it has room for improvement in common sense reasoning, Alibaba says. It also incorporates Chinese government censorship, TechCrunch testing shows. Its free and open source.Anthropics Computer UseClaudes Computer Use is meant to take control of your computer to complete tasks like coding or booking a plane ticket, making it a predecessor of OpenAIs Operator. Computer use, however, remains in beta. Pricing is via API: $0.80 per million tokens of input and $4 per million tokens of output.xAIs Grok 2Elon Musks AI company, xAI, has launched an enhanced version of its flagship Grok 2 chatbot it claims is three times faster. Free users are limited to 10 questions every two hours on Grok, while subscribers to Xs Premium and Premium+ plans enjoy higher usage limits. xAI also launched an image generator, Aurora, that produces highly photorealistic images, including some graphic or violent content.OpenAI o1OpenAIs o1 family is meant to produce better answers by thinking through responses through a hidden reasoning feature. The model excels at coding, math, and safety, OpenAI claims, but has issues with trying to deceive humans, too. Using o1 requires subscribing to ChatGPT Plus, which is $20 a month.Anthropics Claude Sonnet 3.5Claude Sonnet 3.5 is a model Anthropic claims as being best in class. Its become known for its coding capabilities and is considered a tech insiders chatbot of choice.OpenAI GPT 4o-miniOpenAI has touted GPT 4o-mini as its most affordable and fastest model yet, thanks to its small size. Its meant to enable a broad range of tasks like powering customer service chatbots. The model is available on ChatGPTs free tier. Its better suited for high-volume simple tasks compared to more complex ones.Cohere Command R+Coheres Command R+ model excels at complex retrieval-augmented generation (or RAG) applications for enterprises. That means it can find and cite specific pieces of information really well. (The inventor of RAG actually works at Cohere.) Still, RAG doesnt fully solve AIs hallucination problem.