jeffrwells

Nice try Dario.

Alignment is a real and valuable discussion topic. The GP fake tweetstorm is not the correct approach, is my point

I doubt this is a real person. Screams of propaganda. Sama saying GPT-2 is too dangerous to release…all over again.

He joins Twitter for first time in 2026 with a nonsensical username unrelated to his real name, and follows 14 people but is somehow embedded in tech enough to work at Anthropic. I haven’t used twitter since 2014 and even I follow more people.

His morals tell him to walk away from tens of millions in unvested stock due to moral concerns with absolutely no real tangible examples. No reprisals. Fear mongering to juice the stock.

Nice try Dario.

I’ve had a lot of success in the past with fine tuning STT using synthetic data.

I was doing it for Veterinary (ambient recording -> SOAP notes) which has tons of complex domain-specific language AND it is critically important to get right.

“CPR” transcribing as “see pee are” just doesn’t cut it in that industry.

HN highlighting a lot this year, is all

According to ccusage I use about $5k of tokens on my $100/mo Claude Max plan and only hit 5-hr windows where I have to switch tools for a couple hours about once a week

Writing code has never been the hard part.

Writing code has never been the hard part.

Not for Bob, not for you.

In the age of agents, I’ve found the headline claim is even more true

I acquire and operate ecommerce companies, and build a lot of workflows with openclaw-like agents (my own stack).

When it’s working really well, there’s literally no interface needed besides iMessage and email. I’ve built a SaaS app interface style largely to show it off for demos because invisible tools don’t make for great demos

that future is here!

I set up my claw to be able to do that and it’s infinitely better than searching through hours of content myself.

I set it up as experts on topics I already know and like to reduce noise. So I can say “ask [NAME] what I should do about XYZ” and I get back contextualized info and cited video clips

He's not talking about malware awareness. He's talking about a bug i've seen too which requires Claude justifying for *every* tool call to add extra malware-awareness turns. Like every file read of the repo we've been working on

Another way to think about it: every single user of Claude is paying an extra tax in every single request

Reminder that 4.7 may seem like a huge upgrade to 4.6 because they nerfed the F out of 4.6 ahead of this launch so 4.7 would seem like a remarkable improvement...

Anthropic has become a PR vaporware company