Maybe they should up the quality of the various AI features already pushed onto users first? From Google Workspace to the Messages Android app, their AI offerings are half-baked, often useless. My favorite right now is the spellchecking in Google Docs that doesn't detect your language so it autocorrects words that are right over and over again. Doesn't suggest or ask just autocorrects words randomly.

Check who has the cheapest year 1 domain at domcomp. Use Cloudflare pages for free with Bootstrap or roll your own.

One of the most important life lessons I've learnt is to not conflate the model for the reality it tries to capture but also not waste a good model because it doesn't perfectly capture the underlying reality. In this case, poor understanding of both the model and the underlying reality means that people will dismiss both the lean startup methodology and MVP as a technique when both are incredibly powerful tools aimed at exploring if and how a product might be successful.

I've used Mac Whisper to generate closed captions for video clips and it's magical. Saves dozens of hours if not more and there's a free version.

How much would it cost for OpenAI to simulate Reddit?

> What makes this situation so ridiculous is that while we're all watching for scammers attempting to imitate legitimate organisations, FedEx is out there imitating scammers!

Hah!

The executive summary is so well written that a civilian with zero aviation knowledge can understand the incident.

I'm surprised how meh Twitter remained despite Musk's bragging about transforming Twitter into an American WeChat. Like even the recommendation algo for the For You feed is so low quality that it displays the same stuff I just watched in the Following feed. At least try to excel at the engineering parts damnit.