Digg Is Back — This Time as an AI-Powered News Aggregator

Digg keeps coming back. And this time, the resurrection might actually make sense.

Kevin Rose, the original Digg founder who co-acquired the dormant brand last year with Reddit co-founder Alexis Ohanian, has unveiled a completely reimagined version of the site. The new Digg isn't a Reddit clone or a social voting platform — it's an AI-powered news aggregator built specifically around X (formerly Twitter) data, focused on surfacing what's actually worth reading about AI.

TechCrunch confirmed the relaunch today, and the early glimpse suggests a tool that's meaningfully different from anything else in the media landscape right now.

Why Digg Keeps Trying

The original Digg was one of the most important sites on the internet in the mid-2000s — a user-powered news aggregator that could send millions of visitors to any story that hit its front page. Then came the infamous 2010 redesign, a mass exodus to Reddit, and a long decline. The brand was eventually sold, sat dormant, changed hands again, and surfaced as a newsletter-meets-aggregator that eventually shut down in early 2026 after a brief revival.

Rose came back to work on Digg full-time in April. The version he previewed on May 8 looks nothing like its predecessors.

How the New Digg Works

The rebuilt Digg uses real-time ingestion from X to track what's being discussed across the AI community. It runs sentiment analysis, clustering, and signal detection across that stream to figure out what's actually gaining traction — and what's just noise.

The homepage is organized around four featured stories, each representing a different signal:

• Most Viewed: The story generating the most traffic and engagement right now
• Rising Discussion: A story that's picking up momentum fast
• Fastest Climbing: The piece with the sharpest upward trajectory
• In Case You Missed It: Something significant that didn't get the attention it deserved

Below those four features is a ranked list of the day's top stories, complete with engagement metrics — views, comments, likes, and saves — so you can see at a glance why a story is being surfaced.

The goal, per Rose, is to "track the most influential voices in a space" and serve up what's actually worth "paying attention to" in the AI news ecosystem. That's a real problem: there's more AI news published every day than any person could reasonably read, and most of it doesn't matter. The combination of X's real-time social signal with algorithmic ranking could genuinely help.

What Makes This Different From Existing Tools

Most AI news aggregators fall into one of two categories: curated newsletters (edited by humans, often slow and opinionated) or RSS-style feeds (real-time but completely unfiltered). Digg is trying to occupy a different position — automated ranking driven by genuine social signal, not editorial judgment or chronological order.

The reliance on X data is both its strength and its constraint. X is where AI news actually breaks — researchers, founders, journalists, and company accounts all post there first. If you want to know what's happening in AI before it hits the trades, X is where you look. Digg is essentially building an automated, ranked version of what a well-followed AI Twitter list does manually.

The sentiment analysis layer matters too. Not all high-engagement posts are worth reading — controversy drives engagement just as much as genuine significance. Clustering similar posts together prevents any single viral thread from flooding the feed, and signal detection (what Rose calls determining "what matters most") is where the real algorithmic work happens.

The Business Logic

Rose and Ohanian aren't building this purely as a passion project. The longer-term plan is to prove the model on AI news — a community with intense demand for high-quality, low-noise signal — and then expand to other verticals if it works.

That's a sensible approach. AI is arguably the best possible test case: the volume of content is enormous, the audience is global and technically sophisticated, and the signal-to-noise problem is genuinely severe. If an AI-powered aggregator can solve it for AI news, the playbook could extend to climate tech, biotech, finance, or any other information-dense domain with a passionate professional audience.

The X data dependency is worth watching. If X changes its API terms — as it has repeatedly in recent years — Digg's entire data pipeline could become more expensive or restricted overnight. Rose hasn't commented publicly on how he's handling that risk.

What's Next

The version previewed last week is focused exclusively on AI news. Rose has indicated expansion to other topics is planned once the model is validated. No specific timeline has been given for new verticals.

For now, the AI news feed is accessible at Digg's relaunched site. The experience is early — the product has only been live publicly for a matter of days — but the approach is genuinely novel in a media landscape that badly needs better signal extraction.

The Bottom Line

Digg has failed before, and this relaunch carries real risk. But the timing is better than ever. The AI news ecosystem is overwhelming, X remains the dominant platform for real-time technical discourse, and the combination of algorithmic ranking with social signal data is a meaningful differentiator. If Rose and Ohanian have finally figured out what Digg is actually for, this might be the version that sticks.