yeah. people will start finding ways to obfuscate them without harming performance
GENERATIVE AI
-
Inferring Training Distributions from Large Language Models
By
–
if you find this area exciting, there is some fascinating concurrent work which examines the problem of inferring training *distributions* from LLMs: https://
x.com/iamgroot42/sta
tus/1934636691639472630
… (cc @chhaviyadav_ @kamalikac et al!) -
FastDiffusion course teaches comprehensive diffusion model knowledge
By
–
If you finish the fastdiffusion course you'll know a *lot* about diffusion 😀
-
Software Barriers Collapse: AI Disruption Creates 12-18 Month Window
By
–
The pattern here is clear:
→ Software barriers just disappeared
→ Your competition now includes anyone with ChatGPT
→ Most companies are still designing for Software 1.0 The window to build for this new reality is 12-18 months. Opportunities are EVERYWHERE. -
Designing Products for AI Agents as Primary Users
By
–
Build for Agents, Not Just People There's a new category emerging: Humans (use GUIs)
Computers (use APIs)
Agents (use something in between) Companies like @vercel are already designing for AI agents as customers. Your next big user might not be human -
English Replaces Code: Kids Build Apps with Natural Language
By
–
English is the new coding language. Kids are "vibe coding" apps by describing what they want. With things like @FactoryAI you can create entire apps with English words. How magical is that.
-
Iron Man Mental Model: AI Augmentation Over Replacement
By
–
The Iron Man Mental Model AI isn't here to replace humans. It's here to give us superpowers. The spectrum: Augmentation → Agent. Most products will live somewhere in the middle.
-
AI Superpowers: Partial Autonomy Keeps Humans in the Loop
By
–
The Future is Partial Autonomy AI won't replace designers and developers overnight. It will give them superpowers. Smart products will let humans stay in the loop where it matters.
-
LLMs as Simulated Minds: Emergent Psychology and Personality
By
–
LLMs are 'people spirits'. Stochastic simulations of people. They don't just process information. They have emergent psychology, biases, and personality quirks. Like us You're not just prompting a database. You're having a conversation with a simulated mind. This is something
-
LLMs as Operating Systems: The New Computing Paradigm
By
–
LLMs are the New OS Just like you choose Windows vs Mac, you'll choose ChatGPT vs Claude vs Gemini vs Grok based on what 'apps' run best. Distribution will go to whoever builds the most intuitive interface for this new computing paradigm.