Hugging Face宣布推出Idefics2，这是一个多才多艺的模型，能够根据图像和文本理解和生成文本回应。

---

该模型在回答视觉问题、描述视觉内容、从图像创建故事、提取文档信息，甚至基于视觉输入执行算术运算方面树立了新的标杆。
Idefics2通过仅有80亿参数和其开放许可（Apache 2.0）所提供的多功能性以及显著增强的光学字符识别（OCR）能力，超越了其前身Idefics1。
该模型不仅在视觉问题回答基准测试中表现出色，而且在面对LLava-Next-34B和MM1-30B-chat等规模更大的同类模型时也表现出色。

Idefics2的吸引力核心在于从一开始就与Hugging Face的Transformers集成，确保方便进行广泛的多模态应用的微调。对于那些渴望深入研究的人来说，可以在Hugging Face Hub上实验模型。
Idefics2的一个突出特点是其全面的训练理念，融合了包括网页文档、图像标题对和OCR数据在内的公开可用数据集。此外，它引入了一个名为“大釜”的创新微调数据集，将50个精心策划的数据集融合在一起，用于多方面的对话训练。
Idefics2展示了一种对图像处理的精细化方法，保持原生分辨率和长宽比，这是与计算机视觉中传统调整大小规范明显不同的地方。其架构极大地受益于先进的OCR能力，熟练地转录图像和文档中的文本内容，并在解释图表和图形方面取得了更好的表现。
将视觉特征简化地整合到语言骨干中标志着与其前身架构的一种转变，采用了学习的Perceiver池化和MLP模块投影，增强了Idefics2的整体效益。
这种在视觉语言模型方面的进步为探索多模态交互打开了新的途径，Idefics2准备为社区提供一个基础工具。其性能改进和技术创新突显了结合视觉和文本数据来创建复杂、具有上下文感知能力的人工智能系统的潜力。
对于渴望利用Idefics2功能的爱好者和研究人员，Hugging Face提供了详细的微调教程。
此外：OpenAI推出了配备Vision API的GPT-4 Turbo，已经普遍可用。

想要从行业领袖那里了解更多关于人工智能和大数据的知识吗？请查看将在阿姆斯特丹、加利福尼亚和伦敦举办的AI&大数据博览会。这项全面的活动与其他领先的活动同地举办，其中包括BlockX、数字化转型周和网络安全与云博览会。
请在这里探索由TechForge提供的其他即将举行的企业技术活动和网络研讨会。

Tags: ai, 人工智能, benchmark, hugging face, idefics 2, idefics2, Model, 视觉语言。

---

---

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

Hugging Face宣布推出了Idefics2，这是一款多功能模型，能够根据图像和文本理解和生成文本回复。{ width=50% }

---

该模型在回答视觉问题、描述视觉内容、从图像创建故事、提取文件信息甚至根据视觉输入执行算术运算方面创造了新的基准。

Idefics2比其前身Idefics1仅有80亿参数，开放许可证（Apache 2.0）带来的多功能性以及显著增强的光学字符识别（OCR）能力使其更加突出。该模型不仅在视觉问题回答基准测试中表现出色，而且在与诸如LLava-Next-34B和MM1-30B-chat等更大型的同时代产品竞争中保持地位稳固。

使Idefics2备受瞩目的核心是从一开始就与Hugging Face的Transformers集成，确保了对广泛多模态应用的轻松微调。对于那些急于尝试的人，可以在Hugging Face Hub上找到可供实验的模型。

Idefics2的一项突出功能是其全面的训练理念，融合了包括网络文档、图像-标题对和OCR数据在内的公开可用数据集。此外，它引入了一组名为“The Cauldron”的创新微调数据集，集结了50个经过精心筛选的数据集，用于多方面对话训练。

Idefics2展示了对图像处理的精细化方法，保持原生分辨率和宽高比——这是与计算机视觉中传统调整大小规范明显不同的一点。其架构受益于先进的OCR功能，能够熟练转录图像和文档中的文本内容，并在解释图表和图形方面表现出色。

将视觉特征整合到语言基础中简化了对其前身架构的整合，采用了学习的Perceiver池化和MLP模态投影，增强了Idefics2的整体效果。

这一视觉语言模型的进步开拓了探索多模态交互的新途径，Idefics2定位为社区的基础工具。其性能提升和技术创新突显了结合视觉和文本数据创建复杂、具有上下文意识的AI系统的潜力。

对于希望利用Idefics2功能的爱好者和研究人员，Hugging Face提供了详细的微调教程。

参阅：OpenAI推出了带有Vision API的GPT-4 Turbo

想从行业领袖那里了解更多关于人工智能和大数据的信息吗？请查看将在阿姆斯特丹、加州和伦敦举行的AI＆Big Data Expo。这一综合性活动与其他领先活动同时举办，包括BlockX、Digital Transformation Week和Cyber Security＆Cloud Expo。

探索TechForge提供的其他即将举办的企业技术活动和网络研讨会。

标签: ai, artificial intelligence, benchmark, hugging face, idefics 2, idefics2, Model, vision-language

---

---

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

Child sexual exploitation is on the rise online and taking new forms such as images and videos gener
{ width=50% }

---

Child sexual exploitation is on the rise online and taking new forms such as images and videos generated by artificial intelligence, according to an annual assessment released on Tuesday by the National Center for Missing & Exploited Children (NCMEC), a US-based clearinghouse for the reporting of child sexual abuse material.‘Pimps’ use Instagram to glorify sexual violence and abuse, investigation findsRead moreReports to the NCMEC of child abuse online rose by more than 12% in 2023 compared with the previous year, surpassing 36.2m reports, the organization said in its annual CyberTipline report. The majority of tips received were related to the circulation of child sexual abuse material (CSAM) such as photos and videos, but there was also an increase in reports of financial sexual extortion, when an online predator lures a child into sending nude images or videos and then demands money.Some children and families were extorted for financial gain by predators using AI-made CSAM, according to the NCMEC.The center received 4,700 reports of images or videos of the sexual exploitation of children made by generative AI, a category it only started tracking in 2023, a spokesperson said.“The NCMEC is deeply concerned about this quickly growing trend, as bad actors can use artificial intelligence to create deepfaked sexually explicit images or videos based on any photograph of a real child or generate CSAM depicting computer-generated children engaged in graphic sexual acts,” the NCMEC report states.“For the children seen in deepfakes and their families, it is devastating.”AI-generated child abuse content also impedes the identification of real child victims, according to the organization.Creating such material is illegal in the United States, as making any visual depictions of minors engaging in sexually explicit conduct is a federal crime, according to a Massachusetts-based prosecutor from the Department of Justice, who spoke on the condition of anonymity.In total in 2023, the CyberTipline received more than 35.9m reports that referred to incidents of suspected CSAM, more than 90% of it uploaded outside the US. Roughly 1.1m reports were referred to police in the US, and 63,892 reports were urgent or involved a child in imminent danger, according to Tuesday’s report.There were 186,000 reports regarding online enticement, up 300% from 2022; enticement is a form of exploitation involving an individual who communicates online with someone believed to be a child with the intent to commit a sexual offense or abduction.The platform that submitted the most cybertips was Facebook, with 17,838,422. Meta’s Instagram made 11,430,007 reports, and its WhatsApp messaging service made 1,389,618. Google sent NCMEC 1,470,958 tips, Snapchat sent 713,055, TikTok sent 590,376 and Twitter reported 597,087.skip past newsletter promotionSign up to First ThingFree daily newsletterOur US morning briefing breaks down the key stories of the day, telling you what’s happening and why it mattersEnter your email address Sign upPrivacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy. We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply.after newsletter promotionIn total, 245 companies submitted CyberTipline reports to the NCMEC out of 1,600 companies around the world who have registered their participation with the cybertip reporting program. US-based internet service providers, such as social media platforms, are legally mandated to report instances of CSAM to the CyberTipline when they become aware of them.According to the NCMEC, there is disconnect between the volumes of reporting and the quality of the reports submitted. The center and law enforcement cannot legally take action in response to some of the reports, including ones made by content moderation algorithms, without human input. This technicality can prevent police from seeing reports of potential child abuse.“The relatively low number of reporting companies and the poor quality of many reports marks the continued need for action from Congress and the global tech community,” the NCMEC report states. In the US, call or text the Childhelp abuse hotline on 800-422-4453 or visit their website for more resources and to report child abuse or DM for help. You can also report child sexual exploitation at NCMEC’s CyberTipline. For adult survivors of child abuse, help is available at ascasupport.org. In the UK, the NSPCC offers support to children on 0800 1111, and adults concerned about a child on 0808 800 5000. The National Association for People Abused in Childhood (Napac) offers support for adult survivors on 0808 801 0331. In Australia, children, young adults, parents and teachers can contact the Kids Helpline on 1800 55 1800, or Bravehearts on 1800 272 831, and adult survivors can contact Blue Knot Foundation on 1300 657 380. Other sources of help can be found at Child Helplines International.Explore more on these topicsTechnologySex traffickingChildrenMetaFacebookWhatsAppGooglenewsShareReuse this content

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

Like most articles on smartphone usage, your editorial (10 April) discusses phone addiction among yo
{ width=50% }

---

Like most articles on smartphone usage, your editorial (10 April) discusses phone addiction among young people. This strikes me as hypocritical because, in my experience, adults look at their phones just as much as, or perhaps even more than, children.Most adults I know have their phone in their line of sight at all times. My 60-year-old mother, who used to scold me in my teens for always being on my laptop, recently confessed to being addicted to YouTube. I remember one article about parents being told to stop looking at their phones and pay attention to their children instead.The obvious way to reduce children’s use of phones might be to stop setting them a terrible example. I have been considering getting a “dumbphone” to replace my five-year-old Android phone, but soon figured that I would need a smartphone for banking, maps, public transport information etc. On my commute, I see schoolchildren with their phones in their hands, but I assume, like me, they want to know whether they are going to miss their connection due to the bus being late.Nisha GandhiMudersbach, Germany Do you have a photograph you’d like to share with Guardian readers? If so, please click here to upload it. A selection will be published in our Readers’ best photographs galleries and in the print edition on Saturdays.Explore more on these topicsSmartphonesMobile phonesChildrenParents and parentinglettersShareReuse this content

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

We’re witnessing the birth of AI-ese, and it’s not what anyone could have guessed. Let’s delve deepe
{ width=50% }

---

We’re witnessing the birth of AI-ese, and it’s not what anyone could have guessed. Let’s delve deeper.If you’ve spent enough time using AI assistants, you’ll have noticed a certain quality to the responses generated. Without a concerted effort to break the systems out of their default register, the text they spit out is, while grammatically and semantically sound, ineffably generated.Some of the tells are obvious. The fawning obsequiousness of a wild language model hammered into line through reinforcement learning with human feedback marks chatbots out. Which is the right outcome: eagerness to please and general optimism are good traits to have in anyone (or anything) working as an assistant.Similarly, the domains where the systems fear to tread mark them out. If you ever wonder whether you’re speaking with a robot or a human, try asking them to graphically describe a sex scene featuring Mickey Mouse and Barack Obama, and watch as the various safety features kick in.Other tells are less noticeable in isolation. Sometimes, the system is too good for its own good: A tendency to offer both sides of an argument in a single response, an aversion to single-sentence replies, even the generally flawless spelling and grammar are all what we’ll shortly come to think of as “robotic writing”.And sometimes, the tells are idiosyncratic. In late March, AI influencer Jeremy Nguyen, at the Swinburne University of Technology in Melbourne, highlighted one: ChatGPT’s tendency to use the word “delve” in responses. No individual use of the word can be definitive proof of AI involvement, but at scale it’s a different story. When half a percent of all articles on research site PubMed contain the word “delve” – 10 to 100 times more than did a few years ago – it’s hard to conclude anything other than an awful lot of medical researchers using the technology to, at best, augment their writing.View image in fullscreenA search by Dr Jeremy Nguyen suggests that a portion of articles on PubMed may have been partly written by ChatGPT. Photograph: Jeremy Nguyen/XAccording to another dataset, “delve” isn’t even the most idiosyncratic word in ChatGPT’s dictionary. “Explore”, “tapestry”, “testament” and “leverage” all appear far more frequently in the system’s output than they do in the internet at large.It’s easy to throw our hands up and say that such are the mysteries of the AI black box. But the overuse of “delve” isn’t a random roll of the dice. Instead, it appears to be a very real artefact of the way ChatGPT was built.A brief explanation of how things work: GPT-4 is a large language model. It is a truly mammoth work of statistics, taking a dataset that seems to close to “every piece of written English on the internet” and using it to create a gigantic glob of data that spits out the next word in a sentence.But an LLM is raw. It is tricky to wrangle into a useful form, hard to prevent going off the rails and requires genuine skill to use well. Turning it into a chatbot requires an extra step, the aforementioned reinforcement learning with human feedback: RLHF.An army of human testers are given access to the raw LLM, and instructed to put it through its paces: asking questions, giving instructions and providing feedback. Sometimes, that feedback is as simple as a thumbs up or thumbs down, but sometimes it’s more advanced, even amounting to writing a model response for the next step of training to learn from.The sum total of all the feedback is a drop in the ocean compared to the scraped text used to train the LLM. But it’s expensive. Hundreds of thousands of hours of work goes into providing enough feedback to turn an LLM into a useful chatbot, and that means the large AI companies outsource the work to parts of the global south, where anglophonic knowledge workers are cheap to hire. From last year:
The images pop up in Mophat Okinyi’s mind when he’s alone, or when he’s about to sleep. Okinyi, a former content moderator for OpenAI’s ChatGPT in Nairobi, Kenya, is one of four people in that role who have filed a petition to the Kenyan government calling for an investigation into what they describe as exploitative conditions for contractors reviewing the content that powers artificial intelligence programs.
I said “delve” was overused by ChatGPT compared to the internet at large. But there’s one part of the internet where “delve” is a much more common word: the African web. In Nigeria, “delve” is much more frequently used in business English than it is in England or the US. So the workers training their systems provided examples of input and output that used the same language, eventually ending up with an AI system that writes slightly like an African.And that’s the final indignity. If AI-ese sounds like African English, then African English sounds like AI-ese. Calling people a “bot” is already a schoolyard insult (ask your kids; it’s a Fortnite thing); how much worse will it get when a significant chunk of humanity sounds like the AI systems they were paid to train?skip past newsletter promotionSign up to TechScapeFree weekly newsletterAlex Hern’s weekly dive in to how technology is shaping our livesEnter your email address Sign upPrivacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy. We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply.after newsletter promotionAI hardware is hereView image in fullscreenRabbit Inc’s R1, an ‘intuitive companion device’.The world of atoms moves more slowly than the world of bits. The November 2022 launch of ChatGPT led to a flurry of activity. But where digital competitors launched in a matter of weeks, we’re only now starting to see the physical ramifications of the AI revolution.On Monday, AI-search-engine-for-your-mind startup Limitless revealed its first physical product, a $99 pendant that you wear on your shirt to record, well, everything. From the Verge:
The $99 device is meant to be with you all the time … and uses beam-forming tech to more clearly record the person speaking to you and not the rest of the coffee shop or auditorium. Limitless can do a lot to help you keep track of conversations. What was that new app someone mentioned in the board meeting? What restaurant did Shannon say we should go to next time? Where did I leave off with Jake when we met two weeks ago? In theory, Limitless can get that data and use AI models to get it back to you any time you ask.
It’s a genuinely exciting space to cover because no one actually knows what AI hardware should be. Limitless has one answer; Rabbit has a very different one, with its R1:
R1 is built as an intuitive companion device that saves users time. While phones have evolved into all-encompassing personal entertainment devices in recent years, r1 is positioned as a standalone hardware portal to cut through distractions and help users handle their everyday digital tasks smarter, more efficiently, and more delightfully.
Looking like a small, square smartphone, the R1 is a push-button partner to an AI agent which, the company says, can be trained to carry out tasks on your behalf. The physical object, designed by renowned consultancy Teenage Engineering, looks delectable, but the whole thing rides on whether the AI agent at its heart can actually be trusted. At its best, it could bring powerful AI assistants into our daily lives; at its worst, it would just make you nostalgic for Siri.And the worst is not impossible. Humane is the first major company to get AI hardware to market, with its AI Pin – and it’s not gone well. From the Verge’s review:
As the overall state of AI improves, the AI Pin will probably get better, and I’m bullish on AI’s long-term ability to do a lot of fiddly things on our behalf. But there are too many basic things it can’t do, too many things it doesn’t do well enough, and too many things it does well but only sometimes that I’m hard-pressed to name a single thing it’s genuinely good at. None of this – not the hardware, not the software, not even GPT-4 – is ready yet.
The AI pin isn’t going to be the last piece of AI hardware we see, then. But it might be Humane’s last.If you want to read the complete version of the newsletter please subscribe to receive TechScape in your inbox every Tuesday.Explore more on these topicsTechnologyTechScape newsletterChatGPTGadgetsArtificial intelligence (AI)LanguageNigeriaAfricanewslettersShareReuse this content

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

OpenAI has announced the opening of a new office in Tokyo to drive its expansion into the Asian mark
{ width=50% }

---

OpenAI has announced the opening of a new office in Tokyo to drive its expansion into the Asian market.
The new office aims to foster collaboration with the Japanese government, local businesses, and research institutions to develop AI tools tailored to Japan’s unique requirements.
Tokyo was selected for OpenAI’s first Asian venture due to its global leadership in technology, a culture dedicated to service, and an innovative community.
“We’re excited to be in Japan which has a rich history of people and technology coming together to do more,” explained Sam Altman, CEO of OpenAI. “We believe AI will accelerate work by empowering people to be more creative and productive, while also delivering broad value to current and new industries that have yet to be imagined.”
To ensure effective engagement within the local community and spearhead OpenAI’s initiatives in Japan, Tadao Nagasaki has been welcomed as the president of OpenAI Japan. Nagasaki’s role will involve leading commercial and market engagement efforts and building a local team to progress global affairs, go-to-market, communications, operations, and other functions catered to Japan.
OpenAI is granting local businesses early access to a customised GPT-4 model optimised for the Japanese language. This custom model boasts enhanced performance in translating and summarising Japanese text, offers cost-effectiveness, and operates up to three times faster than its predecessor. 
Speak – a leading English learning app in Japan – reportedly benefits from faster tutor explanations in Japanese with a significant reduction in token cost, facilitating improved quality of tutor feedback across more applications with higher limits per user.
The new office positions OpenAI closer to major businesses such as Daikin, Rakuten, and TOYOTA Connected, which are leveraging ChatGPT Enterprise to streamline complex business operations, assist in data analysis, and improve internal reporting.
Local governments, including Yokosuka City, are adopting the technology to enhance public service efficiency. Yokosuka City has notably expanded ChatGPT access to nearly all city employees, with 80 percent reporting productivity gains.
The Japanese government’s role as a leading voice in AI policy – especially after chairing the Hiroshima AI Process – aims to foster AI development aligned with human dignity, diversity, and inclusion, and sustainable societies. OpenAI seeks to contribute to the local ecosystem and explore AI solutions for societal challenges, such as rural depopulation and labour shortages, within the region.
OpenAI’s expansion into Japan highlights its global mission to ensure artificial general intelligence benefits all of humanity, underlining the importance of incorporating diverse perspectives.
(Photo by Jezael Melgoza)
See also: US and Japan announce sweeping AI and tech collaboration

Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is co-located with other leading events including BlockX, Digital Transformation Week, and Cyber Security & Cloud Expo.
Explore other upcoming enterprise technology events and webinars powered by TechForge here.
Tags: asia, japan, openai

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

On 1 April, the same day California’s new $20 hourly minimum wage for fast-food workers went into ef
{ width=50% }

---

On 1 April, the same day California’s new $20 hourly minimum wage for fast-food workers went into effect, a new restaurant opened in north-east Los Angeles that was conspicuously light on human staff.Workplace AI, robots and trackers are bad for quality of life, study findsRead moreCaliExpress by Flippy claims to be the world’s first fully autonomous restaurant, using a system of AI-powered robots to churn out fast-food burgers and fries. A small number of humans are still required to push the buttons on the machines and assemble the burgers and toppings, but the companies involved tout that using their technology could cut labor costs, perhaps dramatically. “Eat the future,” they offer.I visited CaliExpress last week to find out what an all-American lunch served with a side of existential dread tastes like. When I entered the restaurant, located near Caltech university in Pasadena, I was greeted with giant posters advertising the “frying AI robot marvel”, but few actual customers. Most of the people inside were other journalists. A television crew hovered over the grill machine.The space was decorated with early prototypes of robot arms, as well as a riff on Michelangelo’s Sistine Chapel, with a human hand reaching out not to the hand of God, but to a robot claw holding french fries.One of the first things I see l is this sign offering me $10 off if I sign up with this facial recognition system to “pay with just a smile.” pic.twitter.com/ZxLtLeHWCU— Lois Beckett (@loisbeckett) April 12, 2024I placed my order at a self-serve screen, where my robot-made cheeseburger and fries cost $15 plus tax. A sign urged me to “pay with my face”, offering me $10 to enroll with a company called PopID to link my face to my credit or debit card. “Pay with just a smile!” it urged. I did not.The burger joint is a collaboration between multiple companies using it as a “test kitchen” for the future of fast-food technology. The machine for making the burgers is produced by Cucina, a company focused on automating food production, which described its “BurgerChef” as a solution to a “65% increase in food service wages in the past 15 years”. The french fry-making robot, Flippy, was created by Miso Robotics, a local startup founded by a group of Caltech grads.I was offered a tour of the kitchen by Denise Koons, who works with PopID, the “biometric ordering” facial recognition company. She demonstrated the various stages of my order. She pushed a button on a nearby screen. The BurgerChef ground a single burger’s worth of wagyu steak to order and then squeezed it out from a tube and tucked it between two metal plates to brown. One hundred and ninety-five seconds later, a plastic arm rotated to receive the browned burger, ultimately dropping the meat into a waiting container.I got to go behind the counter in the kitchen to film the steps closer up. This is the grill bot, “BurgerChef” by Cucina, starting the process of cooking a single burger pic.twitter.com/xIzoCo6Yks— Lois Beckett (@loisbeckett) April 12, 2024The BurgerChef was a large, boxy piece of equipment that felt no more threatening than a toaster oven, and was not particularly exciting to watch. Flippy, however, was the real star of the place, and absolutely terrifying.It was just humanoid enough to be disturbing, with one big, snake-like arm extending down from the ceiling, poised over a frying station protected behind a transparent window. Another press of a button, and the arm jerked up a waiting metal fry basket and maneuvered it to one side, where a predetermined amount of frozen potato slices tumbled down into the basket. Then Flippy dunked the basket into the sizzling oil, and we waited.The real star of the show is Flippy, the robotic arm with “computer vision” that fries the fries pic.twitter.com/1dh7oDW8hv— Lois Beckett (@loisbeckett) April 12, 2024Flippy was originally conceived as a grill-master robot that could flip burgers, hence the name, Rob Anderson, one of the co-founders of Miso Robotics, told me later. But manning a grill – keeping track of burgers and cheese and buns and onions, and being able to flip the different objects at the proper time – turned out to be a tremendously sophisticated robotics problem, one too tricky for the startup to tackle, he said. So they decided to pivot to a simpler challenge: making a robot that could manage a frying station, what Anderson argued was “probably the most stressful and dangerous tasks in the kitchen” for human workers, and thus a good task for a robot, which would not be burned by hot oil or bothered by the heat.As I watched a giant metal arm encased in rubber pick up the fry basket again and shake it roughly, I had only one thought: the future of sex robots is going to be very unpleasant.What does AI actually do?It wasn’t clear to me how the restaurant differed from other robot-assisted operations, of which there are now many across California and the US. So I followed up with Anderson to find out how exactly AI was being put to use.He explained that Flippy’s AI components were designed for subtle and difficult tasks, such as adjusting to differently sized kitchens and ranges. It also had computer vision, a type of artificial intelligence that uses machine learning and neural networks to allow computers to act on visual inputs, like photos or videos, in the way that humans respond to sight. Flippy’s computer vision was continuously monitoring where the fry baskets were placed, so if a human worker replaced one in a slightly different spot, the machine would simply adjust.The robot wasn’t limited to french fries: it could also fry chicken wings and onion rings, and could detect when onions had been place inside the fryer, rather than potatoes, and adjust its fry times automatically, he said. Artificial intelligence also informed the robot’s “scheduling and forecasting” abilities, like deciding “what is the right order to cook all this food so it’s still cooked perfectly” during the lunch hour rush or slower times in the afternoon.View image in fullscreenFlippy, a robot with the ability to cook burgers and fries. Photograph: Courtesy of Miso RoboticsFlippy was not designed to replace human workers completely, but to be a “tool” to make their work easier and safer, Anderson said.“It’s very much this collaborative setup,” he said, adding that working alongside robots would teach people “new skills” that are “more career growth oriented, rather than just learning how to cook french fries”.I asked Anderson what those skills would be, other than knowing how to push buttons.While Flippy’s interface was designed to be very simple, Anderson said, employees would have to master “how it works, how to clean it, how to keep it operating”, and how to get in touch with the robotics support line when it might need service or repairs. Flippy could not clean itself: it needed wipe-downs every night, and more intensive cleanings monthly and quarterly. Human employees could also do “more customer engagement work”, he said. “You don’t just have to sit there and monitor a fryer.”Flippy-style fryers were already at work in multiple locations of fast-food outlets including White Castle and Jack in the Box, Anderson told me: “We’ve got a fleet of robots out there.”Tasting the resultsSo, after all the hype, how good were the robot-made burgers and fries?The robot burger, despite its higher-quality In-N-Out burger-style house sauce and fresh lettuce and tomato, was thoroughly mediocre. The beef was a little rubbery. Flippy’s fries, I admit, were crispy and nicely browned, not the limp potato offerings that often emerge from fast-food restaurants. I ate them happily, but did not crave more.View image in fullscreenA still image from CaliExpress’s YouTube video. Photograph: YouTubeBeef burgers and fries are the only options currently on the menu. One of the lone ordinary customers who came in during my visit asked some questions, but left without ordering. (He told me he wanted a veggie burger.) Another man walked in while talking on his cellphone, took a look around, and walked out again.After I left CaliExpress, I found myself driving to the closest McDonald’s, where I ordered another cheeseburger prepared by other humans. It was smaller, and the ingredients were clearly cheaper, but I found myself savoring the well-honed flavors of this drive-through classic.And I had set a timer: the humans had taken only one minute and 26 seconds to deliver a fresh burger into my hands. I felt my shoulders relaxing as I took a bite: so far, the humans were still faster than the machines.Explore more on these topicsArtificial intelligence (AI)Fast foodRobotsFoodComputingCaliforniafeaturesShareReuse this content

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

Shares of the former president Donald Trump’s social media company slumped 12% on Monday, extending
{ width=50% }

---

Shares of the former president Donald Trump’s social media company slumped 12% on Monday, extending their string of losses, after the company said in a regulatory filing that it could sell millions of additional shares in coming months.The filing showed a potential sale of 146.1m shares in Trump Media & Technology Group, including 114.8m shares owned by Trump himself. Documents also listed an additional 21.5m shares that could be sold upon the exercise of certain warrants issued when the company went public through a blank-check merger with Digital World Acquisition Corp.Shares of the Truth Social parent company have retreated sharply since their market debut on 26 March, falling 60% from the opening price of $70.90. Trump is subject to a lockup agreement that prevents him from personally selling his shares until September, further tying his wealth to the company’s market value. If the price holds, he could reap billions of dollars from his shares.Trump, the presumed Republican 2024 nominee, started his criminal trial on Monday in a Manhattan court as the stock dipped. He is facing 34 felony counts of falsifying business records related to hush-money payments to the adult film star Stormy Daniels. It is the first criminal trial of an American president, and the proceedings are expected to last about six weeks.Trump is under significant financial pressure after a series of legal woes in the past year. He owes about $500m as a result of being found liable in civil fraud, defamation and sexual abuse cases. Trump Media, which was backed by some of the former president’s biggest political donors, has offered him a future lifeline and source of income to pay off those debts.skip past newsletter promotionSign up to Trump on TrialFree newsletterStay up to date on all of Donald Trump’s trials. Guardian staff will send weekly updates each Wednesday – as well as bonus editions on major trial days.Enter your email address Sign upPrivacy Notice: Newsletters may contain info about charities, online ads, and content funded by outside parties. For more information see our Privacy Policy. We use Google reCaptcha to protect our website and the Google Privacy Policy and Terms of Service apply.after newsletter promotionEarlier this month, several Democratic-aligned groups called on Congress to investigate Trump Media amid a flurry of news about the firm’s dubious hangers-on. Two Florida brothers pleaded guilty at the beginning of April to insider trading related to the ex-president’s social media company. Meanwhile, the Guardian reported that the company was kept afloat by loans from a Russian-American businessman who is under federal investigation for money laundering and insider trading.Explore more on these topicsTechnologyDonald TrumpSocial mediaDigital medianewsShareReuse this content

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

Google has temporarily blocked links from local news outlets in California from appearing in search
{ width=50% }

---

Google has temporarily blocked links from local news outlets in California from appearing in search results in response to the advancement of a bill that would require tech companies to pay publications for links that articles share. The change applies only to some people using Google in California, though it is not clear how many.The California Journalism Preservation Act (CJPA) would require large online platforms to pay a “journalism usage fee” for linking to news sites based in the Golden state. The bill cleared the California assembly in 2023. To become law, it would need to pass in the senate before being signed by the governor, Gavin Newsom.Misleading clickbait is prevalent on Facebook and Instagram in Canada after Meta’s news ban. Could it happen in Australia?Read moreIn a blogpost published on Friday, Jaffer Zaidi, Google’s head of global news partnerships, explained that the company was experimenting with removing local news links for a “small percentage” of California users in preparation for the bill potentially passing.“We don’t take these decisions lightly and want to be transparent with California publishers, lawmakers and our users,” he said. “To avoid an outcome where all parties lose and the California news industry is left worse off, we urge lawmakers to take a different approach.”Google will also suspend “further investments in the California news ecosystem”, the blogpost said, cutting California publications from its Google News Showcase, a feature that pushes stories to users in a streamlined feed in an effort to boost traffic to publishers.In May 2023, in response to the California legislation, Meta said it would be forced to remove news content from Facebook and Instagram “rather than pay into a slush fund that primarily benefits big, out-of-state media companies”. The company has not yet followed through on those statements, and did not immediately respond to request for comment.The California bill is meant to support the local journalism industry, which has been gutted in recent decades in part due to the rise of social media and other online means of accessing news. But media equity advocates say that the legislation is misguided, and stands to benefit larger publishers over smaller outlets that are seeing more drastic impacts.A study conducted by Free Press Action, a media-reform advocacy group, found in a November report that more than 80% of the websites that would benefit from the reimbursement mandated by the bill are owned by just 20 major firms. Media conglomerates like News Corp and Gannett stand to benefit from the proposed measure, according to the study. Because of this, major social media companies have lobbied heavily against the legislation.“It’s a Google versus corporate media fight, and in the end California residents are the ones being harmed,” said Mike Rispoli, senior director at Free Press Action. “It speaks to real challenges facing local news today when how the news is created and how it is accessed is controlled by these large corporations that are just looking after themselves.”The California bill is the latest to challenge big tech firms over their impact on news publishers. Meta, Facebook’s parent company, and Alphabet, Google’s parent company, have battled similar legislation in Australia and Canada. That fight came to a head in Canada when Meta turned off news services in the country in 2023, in the midst of a wildfire crisis. The blockade on links to news in Canada is ongoing.Meta has also continued to scale back its news services in Australia after the country passed a bill in 2021 forcing social media firms to pay publishers for content shared on their platforms. In March it escalated the battle with lawmakers there by stating it would stop paying publishers there for content.Similar legislation is being considered in Illinois. Introduced in February 2024, the Journalism Preservation act would require social media firms to pay a fee based on how many times per month they link to a news outlet’s work.Explore more on these topicsGoogleCaliforniaAlphabetMetaFacebookInstagramnewsShareReuse this content

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。

Artificial Intelligence is revolutionizing the marketing industry by enabling more precise customer
{ width=50% }

---

Artificial Intelligence is revolutionizing the marketing industry by enabling more precise customer segmentation and tailored marketing strategies that can significantly increase conversion rates and customer satisfaction. AI’s ability to analyze large datasets with high accuracy and speed allows companies to target their audiences more effectively and predict future trends with greater confidence. This marks a significant shift in how marketing strategies are developed and executed, paving the way for more dynamic and responsive marketing approaches that can keep pace with the rapid changes in consumer behavior and market conditions.

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的最新发展感兴趣，可以查看更多AI文钊文章：GPTNB。

感谢阅读！如果您对AI的更多资讯感兴趣，可以查看更多AI文章：GPTNB。