Discord is preventing some users from accessing sensitive content unless they allow the platform to verify their age by scanning their face or ID. The new age verification process is described as an “experiment” by Discord, and is being trialed in the United Kingdom and Australia in response to local laws that require online platforms to have guardrails in place for safeguarding children.

The new face and ID scanning requirements only apply to Discord users who haven’t previously verified their age on the platform. Discord says this is a “one-time process” that will be triggered in two situations: when users are exposed to nude or sexually explicit content that’s been flagged by Discord’s sensitive media filter, or when users attempt to change filter control settings to prevent such content from being blurred or blocked. 

An age verification window will then appear that prompts the user to either allow Discord to access their device camera and scan their face, or submit a photograph of their ID by scanning a QR code via their phone.

<img src="https://platform.theverge.com/wp-content/uploads/sites/2/2025/04/Discord-age-verification-scan.jpg?quality=90&strip=all&crop=0,18.285647888672,100,63.428704222657" alt="A screenshot showing a Discord window asking users to submit a face or ID scan to verify their age." title="A screenshot showing a Discord …read more

Wikipedia is attempting to dissuade artificial intelligence developers from scraping the platform by releasing a dataset that’s specifically optimized for training AI models. The Wikimedia Foundation announced on Wednesday that it had partnered with Kaggle — a Google-owned data science community platform that hosts machine learning data — to publish a beta dataset of “structured Wikipedia content in English and French.”

Wikimedia says the dataset hosted by Kaggle has been “designed with machine learning workflows in mind,” making it easier for AI developers to access machine-readable article data for modeling, fine-tuning, benchmarking, alignment, and analysis. The content within the dataset is openly licensed, and as of April 15th, includes research summaries, short descriptions, image links, infobox data, and article sections — minus references or non-written elements like audio files.

The “well-structured JSON representations of Wikipedia content” available to Kaggle users should be a more attractive alternative to “scraping or parsing raw article text” according to Wikimedia — an issue that’s currently putting strain on Wikipedia’s servers as automated AI bots relentlessly consume the platform’s bandwidth. Wikimedia already has content sharing agreements in place with Google and the Internet Archive, but the Kaggle partnership should …read more

Go to any run club in the world and there’s a good chance that everyone there has two things: a Garmin smartwatch to track their run and a Strava account to brag about it. Given the global running boom, it makes Strava’s lack of any modern, in-app training plans a curious and glaring omission. Or, at least, it was until today as Strava is acquiring Runna.

For those who don’t torture themselves with a 6AM daily run, this is big news — even if the companies are keeping mum on the deal’s financial details. Strava is the most well-known fitness social media app on the market. Meanwhile, Runna burst onto the scene in 2021 and has quickly climbed the app charts for folks in need of 5K, 10K, or marathon training plans. Since launch, it’s secured an additional $6.3 million in funding for its AI-powered run coaching, with users spanning 180 countries. In 2024, Runna also tripled the size of its team and is currently hiring roughly 50 roles to expand the product and tech. Peruse running subreddits or RunTok, and you’ll …read more

Towards the end of Mark Zuckerberg’s testimony in a Washington, DC courthouse, a smile flashed across his face. 

Meta’s lead attorney, Mark Hansen, had asked the CEO if he was “happy” about paying $19 billion for WhatsApp in 2014.

Zuckerberg flashed a grin, took a brief pause, and responded, “I’d do it again.”

Over the last decade, the growing scale of Instagram and WhatsApp cemented Meta as one of the most powerful companies on earth. Now, the US government, via a Federal Trade Commission antitrust lawsuit, is trying to unwind those acquisitions through a trial that began on April 14th. Zuckerberg spent roughly 13 hours across three days answering questions from the FTC and Meta lawyers. Much of that time on Wednesday was devoted to attempting to refute one core argument: that he bought both apps to take them out and not to make them better.

The FTC contends that WhatsApp was acquired because Zuckerberg and his executives were worried at the time that private messaging apps would grow into fully-fledged social media businesses. From the witness stand, Zuckerberg acknowledged that it was “something I thought about,” but that he thought it would be “extremel …

Read the full …read more