My Dot reconciled the budget file's remittance section against Wise records, image 1 of 1
Projects

My Dot reconciled the budget file's remittance section against Wise records

added

Here’s me giving my Dot the kind of direction a human coworker might get: “The answer should be somewhere in that spreadsheet.” 🤖😅 I pointed it toward the budget file and its remittance section, but didn’t tell it how to solve the problem. The AI found the relevant data, checked the fees against the Wise records, reconciled the discrepancy, and came back with the answer. Meanwhile, it was also finishing a separate calendar assignment I’d given it earlier. What I especially liked was that, after finding the missing context, it acknowledged that its earlier assumption had been wrong. Increasingly, this feels less like giving a computer instructions and more like giving a capable coworker a clue and letting them take it from there. 🤖🧐

First-hand result from @softypapa (Kurt Bell): he gave his Dot a vague coworker-style direction — “the answer should be somewhere in that spreadsheet” — pointing it at the budget file's remittance section without explaining how to solve it. The Dot found the relevant data, checked fees against Wise records, reconciled the discrepancy, admitted its earlier assumption had been wrong, and meanwhile finished a separate calendar assignment he had given it earlier. Posted with a photo on X.

View on X
See all 95 →
Gave his Dot a real job

Gave his Dot a real job

OpenAI Just Launched Dots. Here's The Truth About It. Get my exact OpenAI Dots setup, safety rules, and copy-paste prompts here: https://sharbel.com/s/8qcs3 OpenAI just launched Dots, an always-on AI agent inside ChatGPT that keeps working when you go offline, messages you first, and uses its own cloud computer to complete ongoing tasks. I got access, created my first Dot, connected it to my tools, and gave it a real job. In this video, I show you how the setup works, what OpenAI genuinely got right, where Dots falls apart, and how it compares with Meta Muse, Grok Bot, Hermes Agent, and OpenClaw. Dots runs on GPT-6 Astra, connects to thousands of apps, learns from your corrections, and can work across ChatGPT, Slack, Teams, and voice calls. Its proactive behavior is the biggest difference. It reads the information you give it access to, identifies potential tasks, and starts surfacing useful work without waiting for another prompt. OpenAI also built in sensible guardrails. Background work is read-only by default, so your Dot can research, draft, and plan, but it needs approval before sending messages, posting publicly, spending money, or taking other risky actions. But there are real limitations. Access is restricted by region and plan, completed tasks consume your usage allowance, and the launch arrived alongside major changes to OpenAI’s paid plans. Dots is also entering a crowded market where free and open-source alternatives already offer many of the same capabilitie
@sharbelxyz