Connect with us

Tech

Netflix is ​​inches closer to getting kicked out of your parent’s account

Published

on

Still couldn’t finish that season of “Stranger Things” you’ve been watching on your ex-roommate’s old Netflix account?

You better hurry: Changes are coming to the streaming platform’s password-sharing policies — and Americans might be the next to get banned.

The Los Gatos-based technology and entertainment company said: press release Wednesday. “Over the past year, we have been exploring different ways to tackle this problem in Latin America, and we are now ready to roll it out more widely in the coming months, starting today in Canada, New Zealand, Portugal and Spain.”

Advertisement

The drastic measures outlined in the Netflix announcement include setting account holders a primary location for their household, then paying the equivalent of four to seven US dollars a month, per profile, to add up to other users who don’t live there.

“A Netflix account is for a single family,” the press release states — a very different framework for the service than the one many subscribers know and love.

A company spokesperson declined to say if or when the policy changes announced Wednesday will make landfall in the United States, only reaffirming the line about a broader rollout soon.

The news shouldn’t come as a surprise to anyone who’s been following the company’s ongoing war against password-sharing software. Netflix said in January Shareholder letter that the company expects to “begin rolling out paid participation more broadly” at some point in the current fiscal quarter.

Password sharing is a widespread problem for Netflix and its competitors in the streaming space Markets appeared To help people find and buy account passwords for cheap.

Advertisement

Losses pile up — one estimate put lost revenue in the account sharing at $25 billion annually, a quarter of which comes from Netflix’s pockets. Amid slowing subscriber growth and downturn across the technology sectorthe pressure on the company to adapt is increasing.

Analysts say that while sharing passwords is somewhat tolerated as a customer acquisition strategy, it eventually begins to limit growth.

“We’re seeing a tipping point starting to have such an impact on subscriber growth, that it’s forcing broadcasters to start taking action,” Ken Gerstein, vice president of sales at anti-piracy consultancy NAGRA, told The Times last April.

In early 2022, Netflix said it was testing features in Chile, Costa Rica and Peru that would allow accounts to pay a little extra to share access with up to two people outside their primary home — the same policy now being implemented in Canada and three other countries.

A change appeared to be imminent last week when information about password policy was tested in Latin America became visible in other countries, which has led to reports that the policy is being publicized more widely.

Advertisement

A company spokesperson said at the time: “For a while… an article was published in the Help Center containing information that only applies to Chile, Costa Rica, and Peru in other countries.” “We’ve updated it since then.”

Now, however, the company’s march against password-sharing software continues far ahead, including in America’s neighbor to the north.

Source link

Advertisement
Continue Reading
Advertisement
Click to comment

Leave a Reply

Your email address will not be published.

Tech

I let the AI ​​pick my makeup for a week

Published

on

By

I Fine artist. Almost every aspect of my life is driven by a desire to create, no matter the medium — from DIY projects to Cosplay and elaborate facial makeupI am constantly making something new. I am always eager to try new technologies, tools and technology, so I am naturally fascinated by AI generators. While I am aware of the ongoing rhetoric surrounding AI art, incl Lawsuits and ethical discussions, my curiosity is much stronger than my apprehension about it.

That’s why I decided to let the AI ​​pick my makeup over the course of five days. For consistency, I used a A dream from Wombo The app to create all the themes featured below. (I also picked this app because there was a 200-character limit per prompt, and I loved the challenge of shorter prompts.) While I did my best to faithfully recreate the look in AI images, I took human liberties based on the supplies I had on hand. And my own hobbies. This is what I made with the help of a machine.



Source link

Advertisement
Continue Reading

Tech

Twitter will only put paid users on your feed

Published

on

By

This comes after a few days Twitter announced Those older verified accounts will lose their blue check mark starting April 1 unless they sign up for the paid Twitter Blue. At the same time, Twitter is working on a method for paid subscribers Hide blue checksprobably because it might seem awkward to have one if all it means is that you paid for it.

Together, both changes could get more subscribers (Twitter hopes), but also ensure that the For You page becomes a collection of shoppers, ramblers, and anyone else who wants to pay for Twitter. Oh, and the brands. By limiting amplification to only a small amount of paid users, it makes the For You page more open, and brands can get more traction and amplification in a free Tweet for paying for Blue than buying ads.

Normal, unpaid accounts are only supposed to be visible in the following feed, the time feed of only people you follow — basically, what Twitter used to be.



Source link

Advertisement

Continue Reading

Tech

We spoke to the man behind the viral photo of the Pope

Published

on

By

Over the weekend, a photo of Pope Francis looking dapper in a white puffer jacket went viral on social media. The 86-year-old seated pope appears to be suffering from some serious cataplexy. But there was just one problem: the photo wasn’t real. Created with Midjourney’s artificial intelligence technical tool.

As word spread across the internet that the image was created by artificial intelligence, many expressed their surprise. “I thought the pope’s puffer jacket was real and never thought about it again,” Chrissy Teigen chirp. “No way can I escape the future of technology.” Garbage Day newsletter writer and former BuzzFeed News correspondent Ryan Broderick invited him “The first real mass-level AI misinformation case,” it follows in the aftermath Fake photos of the arrest of Donald Trump by police in New York last week.

Now, for the first time, the image’s creator has shared the story of how he created the image that fooled the world.

Pablo Xavier, a 31-year-old construction worker from the Chicago area who declined to give his last name due to fears he would be attacked for taking the photos, said he was stumbling through dorm rooms last week when he came up with the idea for the photo.

Advertisement

“I try to figure out ways to make something funny because that’s what I usually try to do,” he told BuzzFeed News. “I try to do funny things or tripartite-psychedelic things. It just dawned on me: I have to do the Pope. Then it came like water: “The Pope in a fluffy Balenciaga coat, Moncler, walking the streets of Rome, Paris, things like that.”

He generated the first three images at around 2pm local time last Friday. (He first started using Midjourney after the death of one of his brothers in November. “It almost all started, just dealing with grief and taking pictures of my ex,” he said. “I fell in love with her after that.”)

When Pablo Xavier first saw the Pope’s photos, he said, “I thought they were perfect.” So he sent it to a Facebook group called AI Art Universe, and then on Reddit. He was shocked when the photos went viral. He said, “I didn’t want it to explode like that.”



Source link

Advertisement
Continue Reading

Trending