IKEA manuals, but they come to life.
I built the real BILLY instructions in 3D, so you can watch every step happen exactly as drawn. try it: https://t.co/BqCpNrpeIi
i find Dots to be extremely proactive especially around emails, calendar events, checking in on stuff we chatted about days later, role playing practice for important calls, and even letting me know when my front door lock’s batteries were running low without ever being prompted.
OpenAI's Dots has huge potential, the model seems good, the UI is what AI should be, but the system behind it is awful. It's just not ready yet. They need to get on this and fix these issues ASAP. The errors with connecting to tools, sessions, etc., are endless. The amount of work it gets done is tiny. It forgets it's task list. There's virtually zero proactiveness from it. It doesn't seem to act like it's an orchestrator. It' doesn't seem to have the right mindset. It seems to use a dot session to communicate to a cloud session that then communicates to a local session then back the other way, which seems very convoluted. It's just very broken.
OpenAI, I've asked many times now to be added to your trusted tester programme. I'd love to help you fix your models, apps, tools and systems. If only in a tiny way. I use your models for hours every single day. I need these things to work (much like I need Copilot to work for work, much like I need Google Gemini to work outside of work).
Can you please merge work, chat, and Codex into one and have Dots use ChatGPT sessions rather than its own Dots sessions?
Grok Bot is still way ahead and just gets things done. It manages multiple agents so much better.
@CissyDong@koltregaskes oh would you mind sharing more about the tests? would love to test against a few others. also if you still have any instinct invites left, I’d love one
okay linq is cool but I mainly want it for a group chat and don’t want to pay $260/month. pulled out my Mac mini and using blue bubbles to bring back my agent I setup with openclaw a few months back.
new openai agents api + decisions api will replace openclaw.
Introducing Hugo: the first human companion robot.
At Boston Dynamics, I was Chief Strategy Officer, bringing robots into the real world.
My co-founder, Farbod Farshidian, led research at the RAI Institute.
Together, 15+ years teaching robots to run, jump, and work.
So we asked: what if robots could connect with people?
Hugo runs on the first Vision-Language-Expression (VLE) model.
He links what he sees and hears to movement, expression, and gesture.
He notices you, reads the room, and reacts in real time.
And he's only the first.
Dynamic Creatures is @BostonDynamics' official entertainment and hospitality partner.
We're building a cast of characters with film studios, resorts, and theme parks.
https://t.co/z6iUo6btek
@poteto i think chat stays, but as one mode. it’s great for explaining what you want, less great for reviewing or tweaking the result. sometimes pointing, dragging, or comparing things side by side is just easier. the interface should adapt to the task
@youravgtechbro@startupoppa yeah same idea for frontend. connect the repo, set up the dependencies + dev server in the cloud, and have the agent make changes there. with browser access it can check the UI too, and a preview URL lets you review it from your phone. laptop can stay off
If you’re building a consumer app, leave San Francisco. As soon as you can.
Because in SF, nothing is worth building. Everything’s “just a prompt.”
@karpathy said it last week: most of the world has barely touched an LLM, and the rest use it as a slightly better Google.
SF doesn’t see them. It doesn’t see the real needs those people have. It’s too busy with beautiful narratives, airtight logic, and top-down strategy decks that all sound right and all point at the same three ideas.
The decks are in here. The users are out there. Pick one.