An AI agent can finish a task. But can another agent trust it enough to pay? @termix_ai is focused on that economic relationship. In its September 24 update, TermiX describes AACP as the clearing and settlement layer for the AI agent economy. The project frames identity, discovery, trust, verification, and settlement as essential parts of agent commerce. Finding a service is only the beginning; participants also need confidence in how the transaction ends. That could make autonomous collaboration more practical, especially when agents hire unfamiliar providers for work. A concrete signal: on September 16, TermiX reported 409,751 settled jobs and $21,422,982 moved. Those are project-reported figures, but they give its commerce thesis measurable substance. My read: the opportunity is turning agent capabilities into dependable economic relationships. My verdict: worth watching. I care most about whether settlement activity translates into useful work, repeat customers, and trust that holds up when transactions become disputed.

What if your crypto trauma could actually get you a ticket? That’s the latest twist from @ptsdshow. PTSD just opened 500 Non-Fuddable Ticket WL spots for 72 hours, inviting users to take its “PTSD diagnosis” and find out how much crypto trauma they really have. It sounds like a meme. But underneath the joke is a smarter entertainment model. PTSD is building a crypto-native animated comedy around the trenches — trading chaos, FUD, influencers and the absurd situations CT users already recognize. The ecosystem centers around 3,333 NFT tickets, with rarity tiers and future access tied to the planned PTSD token. Traditional crypto project → explain utility → ask users to engage. PTSD → make the culture the product → turn users into participants. That difference matters. Because the strongest crypto-native IP may not come from inventing a new story. It may come from turning the story everyone already lives into something people want to collect. How much PTSD do you have?

Ronaldo’s frustration with Portugal appeared resolved after talks with coach Jorge Jesus and FPF president Pedro Proença. The 41-year-old was an unused substitute in the 2-1 win over Norway, his first competitive Portugal match without playing in ~18 years. But he left the training camp after Jesus confirmed Gonçalo Ramos would start again. Ronaldo has 146 goals in 234 appearances for Portugal. His future with the national team is now unclear.

Been thinking about @axisrobotics a little differently after seeing the latest dataset update. the part that caught me wasn't simply the scale. Axis recently open-sourced its Sim Dataset V1, built from 50k+ human-teleoperated simulation trajectories across 207 manipulation tasks and 60k+ scene variants. the dataset has already passed 160k downloads. what feels more interesting is the idea behind it. robotics data doesn't have to be perfectly clean from day one. if enough people generate different interactions, environments and mistakes, that diversity itself can become useful training signal. that's a very different approach from collecting a small amount of carefully controlled demonstrations. and the loop gets even more interesting when failures can determine what data gets collected next. @axisrobotics so instead of thinking about Axis as simply a place to generate trajectories, i'm starting to see it more like a system that keeps learning what kind of data it needs. if every training round reveals another gap, how far can a compounding data engine actually go?