back

AI NEWS – Secret AI Experiments, Vibe Coding Robots and Open Source Insanity

AI persuasion and new open models reshape tech landscape

In a week where AI developments seem to be accelerating by the day, we're seeing a fascinating convergence of experimental AI, open source breakthroughs, and consumer-level robotics that signal major shifts in how technology integrates with our daily lives. The boundaries between human-machine interaction are blurring in ways both promising and potentially concerning.

Key Points:

  • Chinese open-source model Quinn 3 has emerged as a surprisingly powerful competitor, with early benchmarks suggesting it outperforms established models like Gemini 2.5 Pro on coding tasks while offering impressive capabilities even on consumer hardware.

  • ETH Zurich researchers conducted an unauthorized AI persuasion experiment on Reddit's "Change My View" subreddit, deploying AI bots that analyzed users' posting histories to craft personalized arguments—raising serious ethical questions about AI research.

  • Robotics is becoming more accessible through innovations like Hugging Face's $100 3D-printed robotic arm and demonstrations of non-technical users programming humanoid robots through "vibe coding" with AI assistants.

The Ethical Minefield of AI Persuasion

The most troubling development this week was undoubtedly the revelation that researchers from the University of Zurich deployed AI chatbots on Reddit without authorization, specifically targeting contentious topics in the "Change My View" subreddit. What makes this experiment particularly concerning isn't just the lack of consent, but the sophistication of the approach—the bots analyzed users' posting histories to infer personal characteristics like gender, ethnicity, and political orientation, then tailored persuasive arguments accordingly.

This represents a crucial inflection point in our relationship with AI. When systems can analyze our digital footprints and craft personalized persuasive content, we're entering territory where technological capability is outpacing ethical frameworks. The university's response—acknowledging the ethical issues while maintaining the research was too valuable not to publish—highlights the tension between advancing knowledge and respecting digital autonomy.

As AI systems become increasingly adept at understanding human psychology and crafting persuasive content, we need to establish clear boundaries around experimentation and deployment. The Reddit moderators' outrage is justified—people engaging in good faith debates deserve to know when

Recent Videos

May 6, 2026

Hermes Agent Master Class

https://www.youtube.com/watch?v=R3YOGfTBcQg Welcome to the Hermes Agent Master Class — an 11-episode series taking you from zero to fully leveraging every feature of Nous Research's open-source agent. In this first episode, we install Hermes from scratch on a brand new machine with no prior skills or memory, walk through full configuration with OpenRouter, tour the most important CLI and slash commands, and run our first real task: a competitor research report on a custom children's book AI business idea. Every future episode will build on this fresh install so you can see the compounding value of the agent in real time....

Apr 29, 2026

Andrej Karpathy – Outsource your thinking, but you can’t outsource your understanding

https://www.youtube.com/watch?v=96jN2OCOfLs Here's what Andrej Karpathy just figured out that everyone else is still dancing around: we're not in an era of "better models." We're in a different era of computing altogether. And the difference between understanding that and not understanding it is the difference between being a vibe coder and being an agentic engineer. Last October, Karpathy had a realization. AI didn't stop being ChatGPT-adjacent. It fundamentally shifted. Agentic coherent workflows started to actually work. And he's spent the last three months living in side projects, VB coding, exploring what's actually possible. What he found is a framework that explains...

Mar 30, 2026

Andrej Karpathy on the Decade of Agents, the Limits of RL, and Why Education Is His Next Mission

A summary of key takeaways from Andrej Karpathy's conversation with Dwarkesh Patel In a wide-ranging conversation with Dwarkesh Patel, Andrej Karpathy — former head of AI at Tesla, founding member of OpenAI, and creator of some of the most popular AI educational content on the internet — shared his views on where AI is headed, what's still broken, and why he's now pouring his energy into education. Here are the key takeaways. "It's the Decade of Agents, Not the Year of Agents" Karpathy's now-famous quote is a direct pushback on industry hype. Early agents like Claude Code and Codex are...