Forgot how fractions work? Ask an AI assistant: "Explain it like I'm 11. Don't give the answer. Ask me questions." Use it as a tutor, not a cheat sheet. #EverydayAI
@thsottiaux@atelicinvest But what to do if your Dot does not have Computer use ... I feel like a child who received a RC car without batteries for Christmas!
@midotalks@designprejudice I have identified the problem. My Dot has NO Computer use. Any @ that needs it is not available on Dot. I installed Edge, same thing, there is an "Edit a photo" that requires computer use, same thing. They are all available in my normal Codex and Work chats, but not with Dot
@midotalks@designprejudice Ok, I did all of that and in the end it seems there is some glitch. Even Astra later told me exactly what to send to OpenAi via /feedback.
Thank you so much for taking the time to answer though!
@thsottiaux Dot’s settings, all together. My Dot cannot do what my “normal” agents can. I even deleted and started over. I sent a message last night via /feedback.
Parcel text asking for R25? Pause. Remove personal details, codes and the live link before asking AI for warning signs. Verify with the courier independently. AI cannot certify safety. R25 is a South African example.
Ankor's Episode 3: https://t.co/3I6vSfUrEc
What's for supper? List your ingredients, portions, time and equipment. Ask AI for 3 ideas, then check the recipe, food safety and allergy precautions yourself. Episode 2 of Everyday AI from Ankor. Clear inputs, useful options, human review. #EverydayAI
https://t.co/tAbHBW9tHq
@thsottiaux make it possible for Dot to go to older conversations and continue there. I am also struggling with giving it Computer use and Browser use. Its own browser does not pass "Are you human" checks, even if I work in Dot's browser. Dont know if it is just teething problems?
Confusing letter? Remove private details before asking AI what it says, what action it requests and by when. Check dates against the original and confirm with the sender. Ankor Nexa helps organise the follow-up: https://t.co/qbRzcoJ3D3 #EverydayAI
GPT-6.1 Sol is here.
Upgraded with stronger agentic coding and computer use, near-Astra performance, and cached input at a 95% discount to standard input pricing.
GPT-6.1 Sol is built for complex refactors, deep codebase investigations, and long-running agents across apps.