We will know we have reached AGI when AI stops reminding us about our goals and starts asking why we abandoned them. At that point, it is no longer an assistant. It is a life coach with server costs. What is the most passive-aggressive app reminder you have ever received?
@pascal_bornet
-
Demonstrating Model Drift Through Iterative Generative AI Feedback
By
–
For me, this is the best illustration of model drift!
— Pascal Bornet (@pascal_bornet) 11 mai 2026
Ask AI to recreate an image.
Feed the result back in.
Repeat 101 times.
At first, nothing dramatic happens.
Then the details slip.
Faces bend. Objects blur. Reality gets weird.
By the end, it looks like Picasso joined a… pic.twitter.com/432Se38ViLFor me, this is the best illustration of model drift! Ask AI to recreate an image.
Feed the result back in.
Repeat 101 times. At first, nothing dramatic happens.
Then the details slip.
Faces bend. Objects blur. Reality gets weird. By the end, it looks like Picasso joined a -
Lenny’s Buildathon Announcement with Replit Link
By
–
Physics just exposed how easily our eyes can fool us.
— Pascal Bornet (@pascal_bornet) 6 mai 2026
A drone looks completely still inside a moving car.
But here is the twist:
It may look frozen in the air, yet it is actually moving at nearly 100 km/h.
Relative to the car, the drone is still.
Relative to the road, it is… pic.twitter.com/bFL33fJgKGPhysics just exposed how easily our eyes can fool us. A drone looks completely still inside a moving car. But here is the twist: It may look frozen in the air, yet it is actually moving at nearly 100 km/h. Relative to the car, the drone is still.
Relative to the road, it is -
Lenny’s Newsletter x Replit Buildathon Announcement
By
–
𝗧𝗵𝗶𝘀 𝗶𝘀 𝗻𝗼𝘁 𝗷𝘂𝘀𝘁 𝗖𝗵𝗲𝗴𝗴’𝘀 𝗰𝗼𝗹𝗹𝗮𝗽𝘀𝗲. It is a warning to every company built on access to information. Chegg was once worth around $13 billion. Students paid monthly fees to get help with homework, answers, and explanations. Then generative AI changed
-
AI Exposes Both Specialists and Generalists in Career Landscape
By
–
Brutal career truth in the age of AI: knowledge alone is no longer enough. Pure specialists are exposed because AI can learn narrow domains faster than many people can defend them. Shallow generalists are exposed too because LLMs can already summarize, compare, and explain
-
Human-Agent Orchestrator Book Launches to Guide AI-Human Management
By
–
Nine co-authors. Over a hundred contributors. One book. The Human-Agent Orchestrator is out today.
We are the last generation to manage only humans. We wrote the playbook for what comes next, and for right now. I will be honest: this book exists because we got it wrong first. -
AI Tool Access Doesn’t Equal Real Organizational AI Ability
By
–
Professional athletes train most of the time to perform when it matters. Corporate teams perform all the time and get one AI workshop called “transformation.” What worries me is how often we confuse access with ability. Giving people AI tools is not the same as building AI
-
AI Automation and Gen Z Are Reshaping Workplace Expectations
By
–
AI is automating tasks. Gen Z is stress-testing workplace norms.
— Pascal Bornet (@pascal_bornet) 2 mai 2026
This clip is satire, but it points to something real.
What stands out to me is not the bed, the camera, or the contract.
It is the gap between what companies assume and what employees believe they owe.
The future… pic.twitter.com/cEoCKRtcSgAI is automating tasks. Gen Z is stress-testing workplace norms. This clip is satire, but it points to something real. What stands out to me is not the bed, the camera, or the contract.
It is the gap between what companies assume and what employees believe they owe. The future -
Agentic AI Full Stack: Where Most Systems Actually Fail
By
–
Most people still think Agentic AI is just ChatGPT plus tools. They are wrong. This diagram matters because it shows the full stack. Five layers. And in my view, most failures do not happen in the models. They happen in layers four and five. 𝟭/ 𝗔𝗜 & 𝗠𝗟 𝗧𝗵𝗲
-
Fix Your Data: The Hidden Foundation of AI Success
By
–
The best thing companies about AI right now has almost nothing to do with AI. Fix your data. This is the same pattern I keep seeing. Everyone wants the exciting layer:
copilots, agents, demos, dashboards. Almost nobody wants the foundation:
definitions, ownership,