Meta is stuffing its AI chatbot into your Instagram DMs

On Friday, people around the web noticed a new addition to their Instagram: Meta AI, the company’s general-purpose, AI-powered chatbot that can answer questions, write poetry and generate images with a simple text prompt. The move isn’t surprising. Meta revealed Meta AI in September 2023 and has spent the last few months adding the chatbot to products like Facebook Messenger and WhatsApp, so adding it to Instagram seems like a no-brainer. 

“Our generative AI-powered experiences are under development in various phases, and we’re testing a range of them publicly in a limited capacity,” a Meta spokesperson told Engadget, which suggests that not everyone has the feature available yet. TechCrunch, which first noted the news, said that Meta AI was showing up in Instagram’s search bar. But for some of us at Engadget, the feature actually showed up in the search bar in Instagram’s Direct Messaging inbox. 

Tapping it let me start a conversation with Meta AI just I would DM a friend on Instagram. I was able to ask the chatbot to give me definitions of words, suggest headlines for some stories I’m working on, and generate images of dogs on skateboards. I was also able to ask Meta AI to recommend Reels with cats in them, which it was able to do easily.

But when my colleague Aaron Souppouris asked Meta AI in WhatsApp to recommend Reels, it showed him some Reels in that app too — suggesting that the bot in Instagram isn’t really doing anything specific to Instagram. Instead, Meta is simply shoehorning the same chatbot into every app it owns.

If you tap a hamburger menu within the bot, Meta AI will also show you a long list of possible actions you ask the bot to take.

Meta AI will present a list of actions you can ask the bot to take.
Aaron Souppouris

Why would you want a chatbot in Instagram to suggest tips for dealing with credit card debit, have a debate about cardio versus weights, or suggest hacks to travel with points, I do not know. But the point is that if you want to, you can.

This article originally appeared on Engadget at https://www.engadget.com/meta-is-stuffing-its-ai-chatbot-into-your-instagram-dms-231855991.html?src=rss

Google’s new AI video generator is more HR than Hollywood

For most of us, creating documents, spreadsheets and slide decks is an inescapable part of work life in 2024. What's not is creating videos. That’s something Google would like to change. On Tuesday, the company announced Google Vids, a video creation app for work that the company says can make everyone a “great storyteller” using the power of AI.

Vids uses Gemini, Google’s latest AI model, to quickly create videos for the workplace. Type in a prompt, feed in some documents, pictures, and videos, and sit back and relax as Vids generates an entire storyboard, script, music and voiceover. "As a storytelling medium, video has become ubiquitous for its immediacy and ability to ‘cut through the noise,’ but it can be daunting to know where to start," said Aparna Pappu, a Google vice president, in a blog post announcing the app. "Vids is your video, writing, production and editing assistant, all in one."

In a promotional video, Google uses Vids to create a video recapping moments from its Cloud Next conference in Las Vegas, an annual event during which it showed off the app. Based on a simple prompt telling it to create a recap video and attaching a document full of information about the event, Vids generates a narrative outline that can be edited. It then lets the user select a template for the video — you can choose between research proposal, new employee intro, team milestone, quarterly business update, and many more — and then crunches for a few moments before spitting out a first draft of a video, complete with a storyboard, stock media, music, transitions, and animation. It even generates a script and a voiceover, although you can also record your own. And you can manually choose photos from Google Drive or Google Photos to drop them seamlessly into the video.


It all looks pretty slick, but it’s important to remember what Vids is not: a replacement for AI-powered video generation tools like OpenAI’s upcoming Sora or Runway’s Gen-2 that create videos from scratch from text prompts. Instead. Google Vids uses AI to understand your prompt, generate a script and a voiceover, and stitch together stock images, videos, music, transitions, and animations to create what is, effectively, a souped up slide deck. And because Vids is a part of Google Workspace, you can collaborate in real time just like Google Docs, Sheets, and Slides.

Who asked for this? My guess is HR departments and chiefs of staff, who frequently need to create onboarding videos for new employees, announce company milestones, or create training materials for teams. But if and when Google chooses to make Vids available beyond Workspace, which is typically used by businesses, I can also see people using this beyond work like easily creating videos for a birthday party or a vacation using their own photos and videos whenever it becomes available more broadly

Vids will be available in June and is first coming to Workspace Labs, which means you’ll need to opt in to test it. It’s not clear yet when it will be available more broadly.

This article originally appeared on Engadget at https://www.engadget.com/googles-new-ai-video-generator-is-more-hr-than-hollywood-120034992.html?src=rss

Google’s new AI video generator is more HR than Hollywood

For most of us, creating documents, spreadsheets and slide decks is an inescapable part of work life in 2024. What's not is creating videos. That’s something Google would like to change. On Tuesday, the company announced Google Vids, a video creation app for work that the company says can make everyone a “great storyteller” using the power of AI.

Vids uses Gemini, Google’s latest AI model, to quickly create videos for the workplace. Type in a prompt, feed in some documents, pictures, and videos, and sit back and relax as Vids generates an entire storyboard, script, music and voiceover. "As a storytelling medium, video has become ubiquitous for its immediacy and ability to ‘cut through the noise,’ but it can be daunting to know where to start," said Aparna Pappu, a Google vice president, in a blog post announcing the app. "Vids is your video, writing, production and editing assistant, all in one."

In a promotional video, Google uses Vids to create a video recapping moments from its Cloud Next conference in Las Vegas, an annual event during which it showed off the app. Based on a simple prompt telling it to create a recap video and attaching a document full of information about the event, Vids generates a narrative outline that can be edited. It then lets the user select a template for the video — you can choose between research proposal, new employee intro, team milestone, quarterly business update, and many more — and then crunches for a few moments before spitting out a first draft of a video, complete with a storyboard, stock media, music, transitions, and animation. It even generates a script and a voiceover, although you can also record your own. And you can manually choose photos from Google Drive or Google Photos to drop them seamlessly into the video.


It all looks pretty slick, but it’s important to remember what Vids is not: a replacement for AI-powered video generation tools like OpenAI’s upcoming Sora or Runway’s Gen-2 that create videos from scratch from text prompts. Instead. Google Vids uses AI to understand your prompt, generate a script and a voiceover, and stitch together stock images, videos, music, transitions, and animations to create what is, effectively, a souped up slide deck. And because Vids is a part of Google Workspace, you can collaborate in real time just like Google Docs, Sheets, and Slides.

Who asked for this? My guess is HR departments and chiefs of staff, who frequently need to create onboarding videos for new employees, announce company milestones, or create training materials for teams. But if and when Google chooses to make Vids available beyond Workspace, which is typically used by businesses, I can also see people using this beyond work like easily creating videos for a birthday party or a vacation using their own photos and videos whenever it becomes available more broadly

Vids will be available in June and is first coming to Workspace Labs, which means you’ll need to opt in to test it. It’s not clear yet when it will be available more broadly.

This article originally appeared on Engadget at https://www.engadget.com/googles-new-ai-video-generator-is-more-hr-than-hollywood-120034992.html?src=rss

How WhatsApp became the world’s default communication app

In 2014, WIRED asked me to write a few lines about my most-used app as part of an internship application. I wrote about WhatsApp because it was a no-brainer. I was an international student from India, and it was my lifeline to my family and to my girlfriend, now my wife, who lived on the other side of the world. “This cross-platform messenger gets all the credit for my long-distance relationship of two years, which is still going strong,” I wrote in my application. “Skype is great, Google+ Hangouts are the best thing to have happened since Gmail but nothing says ‘I love you’ like a WhatsApp text message.”

A few months into that internship, Facebook announced it was buying WhatsApp for a staggering $19 billion. In WIRED’s newsroom, there were audible gasps at this seemingly minor player's price tag. American journalists weren’t exactly unfamiliar with WhatsApp. But much of the country was still locked in a battle between green and blue bubbles, even as the rest of the world had switched to an app created by two former Yahoo! engineers in WIRED’s Mountain View backyard.

Text messaging was one of the few things you could do on WhatsApp in 2014. There were no emoji you could react with, no high-definition videos you could send, no GIFs or stickers, no read receipts until the end of that year and certainly no voice or video calling. And yet, more than 500 million people around the world were hooked, reveling in the freedom of using nascent cellular data to swap unlimited messages with friends and family instead of paying mobile carriers per text.

WhatsApp’s founders, Jan Koum and Brian Acton, launched the app in 2009 simply to display status messages next to people’s names in a phone’s contact book. But after Apple introduced push notifications on the iPhone later that year, it evolved into a full-blown messaging service. Now, 15 years later, WhatsApp has become a lot more — an integral part of the propaganda machinery of political parties in India and Brazil, a way for millions of businesses to reach customers, a way to send money to people and merchants, a distribution platform for publications, brands and influencers, a video conferencing system and a private social network for older adults. And it is still a great way for long-distance lovers to stay connected.

“WhatsApp is kind of like a media platform and kind of like a messaging platform, but it’s also not quite those things,” Surya Mattu, a researcher at Princeton who runs the university’s Digital Witness Lab, which studies how information flows through WhatsApp, told Engadget. “It has the scale of a social media platform, but it doesn’t have the traditional problems of one because there are no recommendations and no social graph.”

Indeed, WhatsApp’s scale dwarfs nearly every social network and messaging app out there. In 2020, WhatsApp announced it had more than two billion users around the world. It’s bigger than iMessage (1.3 billion users), TikTok (1 billion), Telegram (800 million), Snap (400 million) and Signal (40 million.) It stands head and shoulders above fellow Meta platform Instagram, which captures around 1.4 billion users. The only thing bigger than WhatsApp is Facebook itself, with more than three billion users .

WhatsApp has become the world’s default communications platform. Ten years after it was acquired, its growth shows no sign of stopping. Even in the US, it is finally beginning to break through the green and blue bubble battles and is reportedly one of Meta’s fastest-growing services. As Meta CEO Mark Zuckerberg told the New York Times last year, WhatsApp is the “next chapter” for the company.

Will Cathcart, a longtime Meta executive, who took over WhatsApp in 2019 after its original founders departed the company, credits WhatsApp’s early global growth to it being free (or nearly free — at one point, WhatsApp charged people $1 a year), running on almost any phone, including the world’s millions of low-end Android devices, reliably delivering messages even in large swathes of the planet with suboptimal network conditions and, most importantly, being dead simple, free of the bells and whistles that bloat most other messaging apps. In 2013, a year before Facebook acquired it, WhatsApp added the ability to send short audio messages.

“That was really powerful,” Cathcart told Engadget, “People who don’t have high rates of literacy or someone new to the internet could spin up WhatsApp, use it for the first time and understand it.”

In 2016, WhatsApp added end-to-end encryption, something Cathcart said was a huge selling point. The feature made WhatsApp a black box, hiding the contents of messages from everyone — even WhatsApp — except the sender and the receiver. The same year, WhatsApp announced that one billion people were using the service every month.

That explosive growth came with a huge flip side: As hundreds of millions of people in heavily populated regions, like Brazil and India, came online for the first time, thanks to inexpensive smartphone and data prices, WhatsApp became a conduit for hoaxes and misinformation to flow freely. In India, currently WhatsApp’s largest market with more than 700 million users, the app overflowed with propaganda and disinformation against opposition political parties, cheerleading Narendra Modi, the country’s nationalist Prime Minister accused of destroying its secular fabric.

Then people started dying. In 2017 and 2018, frenzied mobs in remote parts of the country high on baseless rumors about child abductors forwarded through WhatsApp, lynched nearly two dozen people in 13 separate incidents. In response to the crisis, WhatsApp swung into action. Among other things, it made significant product changes, such as clearly labeling forwarded messages — the primary way misformation spread across the service — as well as severely restricting the number of people and groups users could forward content to at the same time.

In Brazil, the app is widely seen as a key tool in the country’s former President Jair Bolsonaro’s 2018 win. Bolsonaro, a far-right strongman, was accused of getting his supporters to circumvent WhatsApp’s spam controls to run elaborate misinformation campaigns, blasting thousands of WhatsApp messages attacking his opponent, Fernando Haddad.

Since these incidents, WhatsApp has established fact-checking partnerships with more than 50 fact-checking organizations globally (because WhatsApp is encrypted, fact-checkers depend on users reporting messages to their WhatsApp hotlines and respond with fact checks). It also made additional product changes, like letting users quickly Google a forwarded message to fact-check it within the app. “Over time, there might be more things we can do,” said Cathcart, including potentially using AI to help with WhatsApp’s fact-checking. “There’s a bunch of interesting things we could do there, I don’t think we’re done,” he said.

Recently, WhatsApp has rapidly added new features, such as the ability to share large files, messages that auto-destruct after they’re viewed, Instagram-like Stories (called Statuses) and larger group calls, among other things. But a brand new feature rolled out globally in fall 2023 called Channels points to WhatsApp’s ambitions to become more than a messaging app. WhatsApp described Channels, in a blog post announcing the launch, as “a one-way broadcast tool for admins to send text, photos, videos, stickers and polls.” They’re a bit like a Twitter feed from brands, publishers and people you choose to follow. It has a dedicated tab in WhatsApp, although interaction with content is limited to responding with emoji — no replies. There are currently thousands of Channels on WhatsApp and 250-plus have more than a million followers each, WhatsApp told Engadget. They include Puerto Rican rapper Bad Bunny (18.9 million followers), Narendra Modi (13.8 million followers), FC Barcelona (27.7 million followers) and the WWE (10.9 million followers). And even though it’s early days, Channels is fast becoming a way for publishers to distribute their content and build an audience.

“It took a year for us to grow to an audience of 35,000 on Telegram,” Rachel Banning-Lover, the head of social media and development at the Financial Times (155,000 followers) told Nieman Lab in November. “Comparatively, we [grew] a similar-sized following [on WhatsApp] in two weeks.”

WhatsApp’s success at consistently adding new functionality without succumbing to feature sprawl has allowed it to thrive, both with its core audience and also, more recently, with users in the US. According to data that analytics firm Data.ai shared with Engadget, WhatsApp had nearly 83 million users in the US in January 2024, compared to 80 million a year before. A couple of years ago, WhatsApp ran an advertising campaign in the US — its first in the country — where billboards and TV spots touted the app’s focus on privacy.

It’s a sentiment shared by Zuckerberg himself, who, in 2021, shared a “privacy-focused vision for social networking” on his Facebook page. “I believe the future of communication will increasingly shift to private, encrypted services where people can be confident that what they say to each other stays secure and their messages and content won’t stick around,” he wrote. “This is the future I hope we will help bring about.”

Meta has now begun using WhatsApp’s sheer scale to generate revenue, although it’s unclear so far how much money, if any, the app makes. “The business model we’re really excited about and one that we’ve been growing for a couple of years successfully is helping people talk to businesses on WhatsApp,” Cathcart said. “That’s a great experience.” Meta monetizes WhatsApp by charging large businesses to integrate the platform directly into existing systems they use to manage interactions with customers. And it integrates the whole system with Facebook, allowing businesses to place ads on Facebook that, when clicked, open directly to a WhatsApp chat with the business. These have become the fastest-growing ad format across Meta, the company told The New York Times.

A few years ago, a configuration change in Facebook’s internal network knocked multiple Facebook services, including WhatsApp, off the internet for more than six hours and ground the world to a halt.

“It’s like the equivalent of your phone and the phones of all of your loved ones being turned off without warning. [WhatsApp] essentially functions as an unregulated utility,” journalist Aura Bogado reportedly wrote on X (then Twitter). In New Delhi and Brazil, gig workers were unable to reach customers and lost out on wages. In London, crypto trades stopped as traders were unable to communicate with clients. One firm claimed a drop of 15 percent. In Russia, oil markets were hit after traders were unable to get in touch with buyers in Europe and Asia placing orders.

Fifteen years after it was created, the messaging app now runs the world.


Engadget 20th anniversary banner

To celebrate Engadget's 20th anniversary, we're taking a look back at the products and services that have changed the industry since March 2, 2004.

This article originally appeared on Engadget at https://www.engadget.com/how-whatsapp-became-the-worlds-default-communication-app-144520113.html?src=rss

How WhatsApp became the world’s default communication app

In 2014, WIRED asked me to write a few lines about my most-used app as part of an internship application. I wrote about WhatsApp because it was a no-brainer. I was an international student from India, and it was my lifeline to my family and to my girlfriend, now my wife, who lived on the other side of the world. “This cross-platform messenger gets all the credit for my long-distance relationship of two years, which is still going strong,” I wrote in my application. “Skype is great, Google+ Hangouts are the best thing to have happened since Gmail but nothing says ‘I love you’ like a WhatsApp text message.”

A few months into that internship, Facebook announced it was buying WhatsApp for a staggering $19 billion. In WIRED’s newsroom, there were audible gasps at this seemingly minor player's price tag. American journalists weren’t exactly unfamiliar with WhatsApp. But much of the country was still locked in a battle between green and blue bubbles, even as the rest of the world had switched to an app created by two former Yahoo! engineers in WIRED’s Mountain View backyard.

Text messaging was one of the few things you could do on WhatsApp in 2014. There were no emoji you could react with, no high-definition videos you could send, no GIFs or stickers, no read receipts until the end of that year and certainly no voice or video calling. And yet, more than 500 million people around the world were hooked, reveling in the freedom of using nascent cellular data to swap unlimited messages with friends and family instead of paying mobile carriers per text.

WhatsApp’s founders, Jan Koum and Brian Acton, launched the app in 2009 simply to display status messages next to people’s names in a phone’s contact book. But after Apple introduced push notifications on the iPhone later that year, it evolved into a full-blown messaging service. Now, 15 years later, WhatsApp has become a lot more — an integral part of the propaganda machinery of political parties in India and Brazil, a way for millions of businesses to reach customers, a way to send money to people and merchants, a distribution platform for publications, brands and influencers, a video conferencing system and a private social network for older adults. And it is still a great way for long-distance lovers to stay connected.

“WhatsApp is kind of like a media platform and kind of like a messaging platform, but it’s also not quite those things,” Surya Mattu, a researcher at Princeton who runs the university’s Digital Witness Lab, which studies how information flows through WhatsApp, told Engadget. “It has the scale of a social media platform, but it doesn’t have the traditional problems of one because there are no recommendations and no social graph.”

Indeed, WhatsApp’s scale dwarfs nearly every social network and messaging app out there. In 2020, WhatsApp announced it had more than two billion users around the world. It’s bigger than iMessage (1.3 billion users), TikTok (1 billion), Telegram (800 million), Snap (400 million) and Signal (40 million.) It stands head and shoulders above fellow Meta platform Instagram, which captures around 1.4 billion users. The only thing bigger than WhatsApp is Facebook itself, with more than three billion users .

WhatsApp has become the world’s default communications platform. Ten years after it was acquired, its growth shows no sign of stopping. Even in the US, it is finally beginning to break through the green and blue bubble battles and is reportedly one of Meta’s fastest-growing services. As Meta CEO Mark Zuckerberg told the New York Times last year, WhatsApp is the “next chapter” for the company.

Will Cathcart, a longtime Meta executive, who took over WhatsApp in 2019 after its original founders departed the company, credits WhatsApp’s early global growth to it being free (or nearly free — at one point, WhatsApp charged people $1 a year), running on almost any phone, including the world’s millions of low-end Android devices, reliably delivering messages even in large swathes of the planet with suboptimal network conditions and, most importantly, being dead simple, free of the bells and whistles that bloat most other messaging apps. In 2013, a year before Facebook acquired it, WhatsApp added the ability to send short audio messages.

“That was really powerful,” Cathcart told Engadget, “People who don’t have high rates of literacy or someone new to the internet could spin up WhatsApp, use it for the first time and understand it.”

In 2016, WhatsApp added end-to-end encryption, something Cathcart said was a huge selling point. The feature made WhatsApp a black box, hiding the contents of messages from everyone — even WhatsApp — except the sender and the receiver. The same year, WhatsApp announced that one billion people were using the service every month.

That explosive growth came with a huge flip side: As hundreds of millions of people in heavily populated regions, like Brazil and India, came online for the first time, thanks to inexpensive smartphone and data prices, WhatsApp became a conduit for hoaxes and misinformation to flow freely. In India, currently WhatsApp’s largest market with more than 700 million users, the app overflowed with propaganda and disinformation against opposition political parties, cheerleading Narendra Modi, the country’s nationalist Prime Minister accused of destroying its secular fabric.

Then people started dying. In 2017 and 2018, frenzied mobs in remote parts of the country high on baseless rumors about child abductors forwarded through WhatsApp, lynched nearly two dozen people in 13 separate incidents. In response to the crisis, WhatsApp swung into action. Among other things, it made significant product changes, such as clearly labeling forwarded messages — the primary way misformation spread across the service — as well as severely restricting the number of people and groups users could forward content to at the same time.

In Brazil, the app is widely seen as a key tool in the country’s former President Jair Bolsonaro’s 2018 win. Bolsonaro, a far-right strongman, was accused of getting his supporters to circumvent WhatsApp’s spam controls to run elaborate misinformation campaigns, blasting thousands of WhatsApp messages attacking his opponent, Fernando Haddad.

Since these incidents, WhatsApp has established fact-checking partnerships with more than 50 fact-checking organizations globally (because WhatsApp is encrypted, fact-checkers depend on users reporting messages to their WhatsApp hotlines and respond with fact checks). It also made additional product changes, like letting users quickly Google a forwarded message to fact-check it within the app. “Over time, there might be more things we can do,” said Cathcart, including potentially using AI to help with WhatsApp’s fact-checking. “There’s a bunch of interesting things we could do there, I don’t think we’re done,” he said.

Recently, WhatsApp has rapidly added new features, such as the ability to share large files, messages that auto-destruct after they’re viewed, Instagram-like Stories (called Statuses) and larger group calls, among other things. But a brand new feature rolled out globally in fall 2023 called Channels points to WhatsApp’s ambitions to become more than a messaging app. WhatsApp described Channels, in a blog post announcing the launch, as “a one-way broadcast tool for admins to send text, photos, videos, stickers and polls.” They’re a bit like a Twitter feed from brands, publishers and people you choose to follow. It has a dedicated tab in WhatsApp, although interaction with content is limited to responding with emoji — no replies. There are currently thousands of Channels on WhatsApp and 250-plus have more than a million followers each, WhatsApp told Engadget. They include Puerto Rican rapper Bad Bunny (18.9 million followers), Narendra Modi (13.8 million followers), FC Barcelona (27.7 million followers) and the WWE (10.9 million followers). And even though it’s early days, Channels is fast becoming a way for publishers to distribute their content and build an audience.

“It took a year for us to grow to an audience of 35,000 on Telegram,” Rachel Banning-Lover, the head of social media and development at the Financial Times (155,000 followers) told Nieman Lab in November. “Comparatively, we [grew] a similar-sized following [on WhatsApp] in two weeks.”

WhatsApp’s success at consistently adding new functionality without succumbing to feature sprawl has allowed it to thrive, both with its core audience and also, more recently, with users in the US. According to data that analytics firm Data.ai shared with Engadget, WhatsApp had nearly 83 million users in the US in January 2024, compared to 80 million a year before. A couple of years ago, WhatsApp ran an advertising campaign in the US — its first in the country — where billboards and TV spots touted the app’s focus on privacy.

It’s a sentiment shared by Zuckerberg himself, who, in 2021, shared a “privacy-focused vision for social networking” on his Facebook page. “I believe the future of communication will increasingly shift to private, encrypted services where people can be confident that what they say to each other stays secure and their messages and content won’t stick around,” he wrote. “This is the future I hope we will help bring about.”

Meta has now begun using WhatsApp’s sheer scale to generate revenue, although it’s unclear so far how much money, if any, the app makes. “The business model we’re really excited about and one that we’ve been growing for a couple of years successfully is helping people talk to businesses on WhatsApp,” Cathcart said. “That’s a great experience.” Meta monetizes WhatsApp by charging large businesses to integrate the platform directly into existing systems they use to manage interactions with customers. And it integrates the whole system with Facebook, allowing businesses to place ads on Facebook that, when clicked, open directly to a WhatsApp chat with the business. These have become the fastest-growing ad format across Meta, the company told The New York Times.

A few years ago, a configuration change in Facebook’s internal network knocked multiple Facebook services, including WhatsApp, off the internet for more than six hours and ground the world to a halt.

“It’s like the equivalent of your phone and the phones of all of your loved ones being turned off without warning. [WhatsApp] essentially functions as an unregulated utility,” journalist Aura Bogado reportedly wrote on X (then Twitter). In New Delhi and Brazil, gig workers were unable to reach customers and lost out on wages. In London, crypto trades stopped as traders were unable to communicate with clients. One firm claimed a drop of 15 percent. In Russia, oil markets were hit after traders were unable to get in touch with buyers in Europe and Asia placing orders.

Fifteen years after it was created, the messaging app now runs the world.


Engadget 20th anniversary banner

To celebrate Engadget's 20th anniversary, we're taking a look back at the products and services that have changed the industry since March 2, 2004.

This article originally appeared on Engadget at https://www.engadget.com/how-whatsapp-became-the-worlds-default-communication-app-144520113.html?src=rss

NVIDIA’s GPUs powered the AI revolution. Its new Blackwell chips are up to 30 times faster

In less than two years, NVIDIA’s H100 chips, which are used by nearly every AI company in the world to train large language models that power services like ChatGPT, made it one of the world’s most valuable companies. On Monday, NVIDIA announced a next-generation platform called Blackwell, whose chips are between seven and 30 times faster than the H100 and use 25 times less power.

“Blackwell GPUs are the engine to power this new Industrial Revolution,” said NVIDIA CEO Jensen Huang at the company’s annual GTC event in San Jose attended by thousands of developers, and which some compared to a Taylor Swift concert. “Generative AI is the defining technology of our time. Working with the most dynamic companies in the world, we will realize the promise of AI for every industry,” Huang added in a press release.

NVIDIA’s Blackwell chips are named in honor of David Harold Blackwell, a mathematician who specialized in game theory and statistics. NVIDIA claims that Blackwell is the world’s most powerful chip. It offers a significant performance upgrade to AI companies with speeds of 20 petaflops compared to just 4 petaflops that the H100 provided. Much of this speed is made possible thanks the 208 billion transistors in Blackwell chips compared to 80 billion in the H100. To achieve this, NVIDIA connected two large chip dies that can talk to each other at speeds up to 10 terabytes per second.

In a sign of just how dependent our modern AI revolution is on NVIDIA’s chips, the company’s press release includes testimonials from eight CEOs who collectively lead companies worth trillions of dollars. They include OpenAI CEO Sam Altman, Microsoft CEO Satya Nadella, Alphabet CEO Sundar Pichai, Meta CEO Mark Zuckerberg, Google DeepMind CEO Demis Hassabis, Oracle chairman Larry Ellison, Dell CEO Michael Dell, Amazon CEO Andy Jassy and Tesla CEO Elon Musk.

“There is currently nothing better than NVIDIA hardware for AI,” Musk says in the statement. "Blackwell offers massive performance leaps, and will accelerate our ability to deliver leading-edge models. We’re excited to continue working with NVIDIA to enhance AI compute,” Altman says.

NVIDIA did not disclose how much Blackwell chips would cost. Its H100 chips currently run between $25,000 and $40,000 per chip, according to CNBC, and entire systems powered by these chips can cost as much as $200,000.

Despite their costs, NVIDIA’s chips are in high demand. Last year, delivery wait times were as high as 11 months. And having access to NVIDIA’s AI chips is increasingly seen as a status symbol for tech companies looking to attract AI talent. Earlier this year, Zuckerberg touted the company’s efforts to build “a massive amount of infrastructure” to power Meta’s AI efforts. “At the end of this year,” Zuckerberg wrote, “we will have ~350k Nvidia H100s — and overall ~600k H100s H100 equivalents of compute if you include other GPUs.”

This article originally appeared on Engadget at https://www.engadget.com/nvidias-gpus-powered-the-ai-revolution-its-new-blackwell-chips-are-up-to-30-times-faster-001059577.html?src=rss

NVIDIA’s GPUs powered the AI revolution. Its new Blackwell chips are up to 30 times faster

In less than two years, NVIDIA’s H100 chips, which are used by nearly every AI company in the world to train large language models that power services like ChatGPT, made it one of the world’s most valuable companies. On Monday, NVIDIA announced a next-generation platform called Blackwell, whose chips are between seven and 30 times faster than the H100 and use 25 times less power.

“Blackwell GPUs are the engine to power this new Industrial Revolution,” said NVIDIA CEO Jensen Huang at the company’s annual GTC event in San Jose attended by thousands of developers, and which some compared to a Taylor Swift concert. “Generative AI is the defining technology of our time. Working with the most dynamic companies in the world, we will realize the promise of AI for every industry,” Huang added in a press release.

NVIDIA’s Blackwell chips are named in honor of David Harold Blackwell, a mathematician who specialized in game theory and statistics. NVIDIA claims that Blackwell is the world’s most powerful chip. It offers a significant performance upgrade to AI companies with speeds of 20 petaflops compared to just 4 petaflops that the H100 provided. Much of this speed is made possible thanks the 208 billion transistors in Blackwell chips compared to 80 billion in the H100. To achieve this, NVIDIA connected two large chip dies that can talk to each other at speeds up to 10 terabytes per second.

In a sign of just how dependent our modern AI revolution is on NVIDIA’s chips, the company’s press release includes testimonials from eight CEOs who collectively lead companies worth trillions of dollars. They include OpenAI CEO Sam Altman, Microsoft CEO Satya Nadella, Alphabet CEO Sundar Pichai, Meta CEO Mark Zuckerberg, Google DeepMind CEO Demis Hassabis, Oracle chairman Larry Ellison, Dell CEO Michael Dell, Amazon CEO Andy Jassy and Tesla CEO Elon Musk.

“There is currently nothing better than NVIDIA hardware for AI,” Musk says in the statement. "Blackwell offers massive performance leaps, and will accelerate our ability to deliver leading-edge models. We’re excited to continue working with NVIDIA to enhance AI compute,” Altman says.

NVIDIA did not disclose how much Blackwell chips would cost. Its H100 chips currently run between $25,000 and $40,000 per chip, according to CNBC, and entire systems powered by these chips can cost as much as $200,000.

Despite their costs, NVIDIA’s chips are in high demand. Last year, delivery wait times were as high as 11 months. And having access to NVIDIA’s AI chips is increasingly seen as a status symbol for tech companies looking to attract AI talent. Earlier this year, Zuckerberg touted the company’s efforts to build “a massive amount of infrastructure” to power Meta’s AI efforts. “At the end of this year,” Zuckerberg wrote, “we will have ~350k Nvidia H100s — and overall ~600k H100s H100 equivalents of compute if you include other GPUs.”

This article originally appeared on Engadget at https://www.engadget.com/nvidias-gpus-powered-the-ai-revolution-its-new-blackwell-chips-are-up-to-30-times-faster-001059577.html?src=rss

TikTok’s CEO urges users to ‘protect your constitutional rights’ as US ban looms

Hours after the House passed a bill that could ban TikTok in the United States, Shou Chew, the company’s CEO urged users to “protect your constitutional rights.” Chew also implied that TikTok would mount a legal challenge if the bill is passed into law.

“We will not stop fighting and advocating for you,” Chew said in a video posted to X. “We will continue to do all we can including exercising our legal rights to protect this amazing platform that we have built with you.” He also asked TikTok users in the US to share their stories with friends, families, and senators. “This legislation, if passed into law, will lead to a ban of TikTok in the United States,” Chew said. “Even the bill’s sponsors admit that’s their goal.”

The bill, known as the “Protecting Americans from Foreign Adversary Controlled Applications Act” passed the House on Wednesday with bipartisan support just days after it was introduced. Should the bill pass into law, it would force TikTok’s parent company ByteDance, a Chinese corporation, to sell TikTok to a US company within six months, or be banned from US app stores and web hosting services. TikTok has challenged state-level bans in the past. Last year, TikTok sued Montana, which banned the app in the state. A federal judge temporarily blocked that ban in November before it went into effect.

Last week, TikTok sent push notifications to the app’s more than 170 million users in the US urging them to call their representatives about the potential ban. “Speak up now — before your government strips 170 million Americans of their Constitutional right to free expression,” the notification said. The wave of notifications reportedly led to House staffers being inundated with calls from high schoolers asking what a Congressman is. Lawmakers criticized the company they perceived as trying to “interfere” with the legislative process.

In his appeal, Chew said that banning TikTok would give “more power to a handful of other social media companies.” Former President Donald Trump, who once tried to force ByteDance to sell TikTok in the US, recently expressed a similar sentiment, claiming that banning TikTok would strengthen Meta whose platform, Reels, competes with TikTok directly. Chew also added that taking TikTok away would also hurt hundreds of thousands of American jobs, creators, and small businesses.

This article originally appeared on Engadget at https://www.engadget.com/tiktoks-ceo-urges-users-to-protect-your-constitutional-rights-as-us-ban-looms-002806276.html?src=rss

TikTok’s CEO urges users to ‘protect your constitutional rights’ as US ban looms

Hours after the House passed a bill that could ban TikTok in the United States, Shou Chew, the company’s CEO urged users to “protect your constitutional rights.” Chew also implied that TikTok would mount a legal challenge if the bill is passed into law.

“We will not stop fighting and advocating for you,” Chew said in a video posted to X. “We will continue to do all we can including exercising our legal rights to protect this amazing platform that we have built with you.” He also asked TikTok users in the US to share their stories with friends, families, and senators. “This legislation, if passed into law, will lead to a ban of TikTok in the United States,” Chew said. “Even the bill’s sponsors admit that’s their goal.”

The bill, known as the “Protecting Americans from Foreign Adversary Controlled Applications Act” passed the House on Wednesday with bipartisan support just days after it was introduced. Should the bill pass into law, it would force TikTok’s parent company ByteDance, a Chinese corporation, to sell TikTok to a US company within six months, or be banned from US app stores and web hosting services. TikTok has challenged state-level bans in the past. Last year, TikTok sued Montana, which banned the app in the state. A federal judge temporarily blocked that ban in November before it went into effect.

Last week, TikTok sent push notifications to the app’s more than 170 million users in the US urging them to call their representatives about the potential ban. “Speak up now — before your government strips 170 million Americans of their Constitutional right to free expression,” the notification said. The wave of notifications reportedly led to House staffers being inundated with calls from high schoolers asking what a Congressman is. Lawmakers criticized the company they perceived as trying to “interfere” with the legislative process.

In his appeal, Chew said that banning TikTok would give “more power to a handful of other social media companies.” Former President Donald Trump, who once tried to force ByteDance to sell TikTok in the US, recently expressed a similar sentiment, claiming that banning TikTok would strengthen Meta whose platform, Reels, competes with TikTok directly. Chew also added that taking TikTok away would also hurt hundreds of thousands of American jobs, creators, and small businesses.

This article originally appeared on Engadget at https://www.engadget.com/tiktoks-ceo-urges-users-to-protect-your-constitutional-rights-as-us-ban-looms-002806276.html?src=rss

Elon Musk kills Don Lemon’s new X show before it ever began

X has canceled a high-profile partnership with former CNN host Don Lemon to stream a video talk show on the platform. Lemon said that the company canceled his contract hours after he interviewed X’s billionaire owner Elon Musk for the first episode of “The Don Lemon Show,” which was scheduled to stream on the platform this Monday.

“Elon Musk is mad at me,” Lemon said in a video posted to X on Wednesday. “Apparently, free speech absolutism doesn’t apply when it comes to questions about him from people like me.”

Lemon’s announcement came a day after company CEO Linda Yaccarino declared that X was becoming a “video first” platform. It announced the partnership with Lemon in January as part of a larger strategy to stream more original content on the service. This included striking deals with former representative Tulsi Gabbard and sports radio commentator Jim Rome to stream their own shows on the platform. Last year, X reportedly made a similar deal with Tucker Carlson after he was fired from his hosting duties at Fox News. X’s decision to cancel Lemon’s show raises questions about the company’s strategy.

“The Don Lemon Show is welcome to publish its content on X, without censorship, as we believe in providing a platform for creators to scale their work and connect with new communities,” X said in a statement. “However, like any enterprise, we reserve the right to make decisions about our business partnerships, and after careful consideration, X decided not to enter into a commercial partnership with the show.”

Lemon said that he will now stream the first episode of “The Don Lemon Show” on X, YouTube and other podcast platforms, and is preparing for a legal fight in case X refuses what is reportedly a multi-million dollar payout. “Don has a deal with X and he expects to be paid for it,” a spokesperson for Lemon told Variety. “If we have to go to court, we will.” However, two anonymous sources claimed to Semafor that Lemon may not have actually signed a contract with X. Musk has a history of withholding payments. A group of former Twitter executives including the company’s ex-CEO Parag Agrawal are suing Musk and X over millions of dollars in unpaid severance benefits.

Lemon’s interview with Musk, which was recorded on Friday, spanned a wide range of topics including the presidential election, and, reportedly, the billionaire’s alleged ketamine use, the subject of a Wall Street Journal story published earlier this year. “Hardcore questions were asked,” Lemon told an X user. In a written statement, Lemon said that he had a “good conversation” with Musk, but the billionaire clearly didn’t seem to think so.

Musk wrote that Lemon’s approach was “basically just ‘CNN, but on social media’, which doesn’t work, as evidenced by the fact that CNN is dying,” in response to a user asking X about specific reasons for terminating the partnership with Lemon. “And, instead of it being the real Don Lemon," Musk sniped, "it was really just [former CNN President] Jeff Zucker talking through Don, so lacked authenticity.”

Lemon was fired from CNN nearly a year ago after making on-air remarks against former Republican presidential contender Nikki Haley that many considered sexist and ageist, as well as reports showing he engaged in misogynistic behavior over his 15-year tenure at CNN.

This article originally appeared on Engadget at https://www.engadget.com/elon-musk-kills-don-lemons-new-x-show-before-it-ever-began-205608734.html?src=rss