@patife@HenrM_Cruz I also don't understand. The privacy and local-first angle might explain it but it should be a non-issue on Macs.
Even Google went for it at Android scale: https://t.co/V3JlGp2B7D
@WisprFlow & co are right there for the taking. Maybe all the AI people were fixing Siri?
Somewhere out there is a guy who uses Notion, Superhuman, OpenClaw on a Mac Mini, Raycast, a mechanical keyboard ($400), Wispr Flow, and gets nothing done every day
@HamelHusain@modemdev Particularly if Fast Mode is enabled. No longer feels like watching a movie in slow motion and like something I'd do faster myself.
The reason AI is so good at coding is not just the models. The people building coding agents use them all day, every day. That feedback loop compounds.
@amix3k I agree and wouldn't bet against it as well.
But the X/VC bubble makes it seem we are a few months away from both technical feasibility and commercial viability, which seems an exaggeration.
Companies bending physics always reminds me of https://t.co/N1hhmnGEIw
@amix3k I think his point is about the volume and magnitude of breakthroughs that need to happen for this to become viable, which the current narrative around the topic glosses over.
At some point you're battling some pretty fundamental physics laws—which he explains well.
@amix3k After seeing that last screenshot of what appears to be Twist: would love an update on how it has been going!
(Daniel here, we spoke briefly at the end of Lisbon AI but the conversation didn't get there!)
@rikrenard This was actually a decent product spec, so I vibe-coded it for you: https://t.co/SDOZxy76OF
(has basic text to speech and .fax file download)