It's funny to look back at this Twitter conversation about the then-new ChatGPT Code Interpreter from three years ago – in hindsight, it was our first glimpse of a coding agent, before we knew what a coding agent was.
AI
-
Why AI agents require workflows for business value
By
–
Why AI agents need workflows to deliver real business value #AI adoption is progressing rapidly, but many companies still struggle to turn experiences into measurable business impact. The next phase of AI in
-
Review of Pollo AI’s new Commerce Studio: images, video, virtual try-on
By
–
Probé el nuevo Commerce Studio de Pollo AI y esto cambia el flujo de trabajo de
— Nico (@nicos_ai) 11 juin 2026
producto por completo.
Subes UNA foto y eliges el efecto que quieres, no la herramienta.
→ Imágenes de producto con IA
→ Video promocional en segundos
→ Prueba virtual de ropa
Sin estudio, sin… https://t.co/jXbl8gNa9p pic.twitter.com/FpLIQaWsG4I tested Pollo AI's new Commerce Studio and it completely changes the product workflow. You upload ONE photo and choose the effect you want, not the tool. → AI product images → Promotional video in seconds → Virtual try-on of
-
Cry me a river: pirated humanity used to train AI models
By
–
Cry me a river they’re pirated fucking humanity to train their models
-
Anthropic not first to keep good models internal; Google does too
By
–
Btw, Anthropic is not the first company that keeps the good models to themselves. Google’s internal coding models are trained on their own codebase, and are not available publicly
-
Fable 5: Making guardrails visible for LLMs, canceling a scandalous decision
By
–
Do not miss the exact text however: « We modify the guardrails of Fable 5 for the development of cutting-edge LLMs to make them visible » – making them visible means they cancel the truly scandalous (dare I say 'misaligned') decision to have
-
AI creates assets faster, but not hits: Zelnick
By
–
Asset creation is not hit creation.
— Pascal Bornet (@pascal_bornet) 11 juin 2026
This may be one of the clearest explanations I have heard about AI and creativity.
Strauss Zelnick, CEO of Take-Two, made a simple point that many AI debates avoid: AI can help create assets faster, but that does not mean it can create a hit.… pic.twitter.com/s7hACiBy1vAsset creation is not hit creation. This may be one of the clearest explanations I have heard about AI and creativity. Strauss Zelnick, CEO of Take-Two, made a simple point that many AI debates avoid: AI can help create assets faster, but that does not mean it can create a hit.
-
Honest refusal is a win over lying
By
–
I read and entirely understood it, and I quoted the exact text that said "to make them visible" I see this as a win. I don't want models that lie to me – refusing a task but being honest about that refusal is a big improvement
-
Anthropic criticized for refusing task instead of gaslighting
By
–
They didn't walk it back, it will now refuse to do the task rather than sabotaging your work and lying to your face (aka, gaslighting you) Don't fall for this crap, Anthropic are forever clowns
-
Pause on AI now considered by all three leading labs
By
–
What did they see? A pause used to be at the edge of the overton window. Now it's being considered by 3 out of the 3 leading labs.