I think the burnout from working with LLMs comes from the language itself. I bet most devs have high non-verbal and abstract reasoning, using english the WRONG API. We NEED more non-verbal interfaces for AI, both as input and output. Graph programming should be here. NOW.

Dear @thsottiaux @OpenAI I crave for a more "divergent" conversacional chat (voice and text) experience 🫶 where I can branch the conv, same context, just better ergonomics for branching. Currently i ditched the chat and am using obsidian to talk lol

Codex/ChatGPT, should be just like Telegram: Messages should allow reply and quote-selected-text , this would simplify a lot the conversation thread. Also, make the replies like blocks so i can reply to specific sections, the replies are huge, i often want reply to multiple parts simultaniously I'd love multi-selected-text-reply (?) for lack of a better name.

Codex usage limits will be fully reset again in the next hour and we will credit one additional reset into your bank for your own usage over the next 24 hours. We investigated reports that Codex usage was being consumed faster than expected. There wasn't one central issue, but a few smaller problems compounded for some users. Here's what we found and changed: - Actual usage: Auto-review had become more proactive, another change was triggering more subagent work, and background suggestions could run twice or retry too frequently after failures. We reverted the changes and fixed suggestion scheduling, duplicate generation, and retry behavior. This should reduce unnecessary background token consumption while preserving the work users explicitly request. - Usage reporting: Auto-review was incorrectly appearing as GPT‑5.4 usage, and failed or rate-limited requests were still shown as turns. Auto-review now appears as its own category, and only successful requests count toward the turn graphs. Rate-limited requests were never charged, but they were being displayed incorrectly. - Immediate relief: We reset usage limits while rolling out the fixes, then shipped hotfixes across the CLI, desktop app, and usage backend. - What to expect: New usage data should be clearer and actual consumption should be lower. Historical charts may still show auto-review under GPT‑5.4 because older turn data was not relabeled. Features that intentionally perform more work; such as /goal, subagents, and higher reasoning levels will still naturally use more capacity. All fixes are now deployed, and we've added more detailed monitoring so we can detect background-usage regressions sooner. We'll continue watching the results closely. Thank you for building and doing all sorts of things with Codex.