Dead internet theory has become reality!

Dear Readers,

Have you noticed how the web feels… quieter lately? Not in volume, but in pulse. Scroll long enough and you’ll feel it, the eerie sameness, the missing spark of a human voice. Reddit’s Alexis Ohanian calls it “the dead internet,” a realm flooded with bots and AI slop that mimic us but never quite are us. The next frontier of AI isn’t just about smarter models - it’s about whether we can still tell who’s real in the digital crowd.

In today’s issue, we explore this uneasy threshold. From OpenAI’s new Well-Being Council shaping safer AI behavior, to Alibaba’s compact Qwen3-VL models redefining efficiency, and a striking call for authenticity in online life, today’s stories trace the fight to keep meaning alive in a machine-driven world. Let’s find out what’s still human on the web, and how we might keep it that way.

In Today’s Issue:

💀 Alexis Ohanian warns the "dead internet theory" is a reality

📄 A new paper warns of "survey paper DDoS" from AI-generated research

🧠 Researchers unveil a hierarchical reasoning model

🖼️ VIRTUE allows visual-interactive text-image embedding

✨ And more AI goodness…

All the best,

Alibaba introduces smaller 4B and 8B Qwen3-VL models that retain full multimodal capabilities while slashing VRAM demands. Despite their reduced size, these models outperform Gemini 2.5 Flash Lite and GPT-5 Nano in areas like STEM reasoning, OCR, video understanding, and agent tasks—sometimes even rivaling the 72B flagship from six months ago. FP8 versions further optimize deployment efficiency.

Learn More ►

OpenAI plans to loosen ChatGPT’s restrictions after addressing major mental health concerns, making it more natural, expressive, and customizable again. Starting in a few weeks, users can choose more human-like personalities, and verified adults will gain broader freedoms—including access to adult content by December.

The Takeaway

👉 Alexis Ohanian warns that much of today’s internet is dominated by bots and AI-generated content.

👉 He argues platforms must restore authenticity through “proof of life” systems verifying real humans.

👉 The AI boom has blurred the line between organic and synthetic engagement, eroding user trust.

👉 The next evolution of the web will hinge on rebuilding credibility and real human connection.

It’s a chilling thought: “much of the internet is now dead.” That’s the warning Alexis Ohanian, Reddit Co-founder, recently issued, and it’s pushing every AI-enthusiast to sit up and ponder our digital future.

Ohanian claims that large swathes of what we experience online-LinkedIn, Twitter, comment sections-aren’t driven by humans anymore, but by bots or semi-automated AI “slop.”

This idea connects directly with the Dead Internet Theory, which argues that most social media activity is now non-human in origin.

In Ohanian’s vision, the cure lies in “proof of life”-platforms that can reliably verify human presence and rebuild trust.

The future might belong to social spaces built on human verification, or to group chats where trust is intrinsic. Will we let the internet become a hollow arena, or revive it from within?

Why it matters: This conversation reframes the battle for the soul of the internet: if humanity loses its voice online, our digital world becomes sterile and hollow. We, the creators and users of AI, hold a key role in reclaiming spaces of real dialogue.

Sources:

🔗 https://www.businessinsider.com/alexis-ohanian-much-of-the-internet-is-now-dead-2025-10

This paper warns that AI-generated survey papers are flooding platforms like arXiv, overwhelming readers and distorting scientific value. The authors introduce the idea of a “survey paper DDoS” and propose curated, versioned community reviews as a solution.

Learn More ►