I feel the same. In fact, the less skilled I am in something, I’ve found it to be much more useful to shield myself from LLM usage for a while, until I feel fluent in it. And only then start delegating to LLM. Otherwise, the process of correcting mistakes I barely understand becomes too inefficient.

If you are worried about disinformation and media manipulation in this election give a pass to
@ctvyourmorning.bsky.social
They caved to pressure from Pierre Poiievre's online posse in their continual harassment of @rachelgilmore.bsky.social
She is a truth teller.
CTV is not.

This feels like a direct response to the Honey extension scandal!

Used OpenAI's Whisper to dictate a few sentences. Pardon the potentially dumb, and not-so-well-phrased question, but I asked the question and stopped. Then Whisper added a "Thank you for watching" at the end!

This is far and away the most crazy hallucination I've experienced first hand so far!

Yesterday’s live is now on YouTube: m.youtube.com/watch?featur...

To be re-evaluated again next month. Hoping things will have gone back to normal by then, but fully prepared to go ahead with the cancellation.

That feeling when you accidentally end up testing non-functional aspects of your app (load time) because you had some end-to-end tests that you just couldn't figure out how to write other than relying on timers 😅

(P.S. found the offending commit introducing the slowdown and modified it! 🎊)

Dear Netflix, can you have the civility of allowing me to breathe for 5 seconds after a show I love ends in a dramatic and emotional way, to maybe listen to the music and digest what I just saw before shoving the next output of your recommendation algorithm in my face? 😔