chrisjj

22 ms·

841.
by chrisjj 7mo ago
Nor factory reset it. Reports suggest this will now brick the unit.
842.
by chrisjj 7mo ago
Anti-US sentiment had nothing to do with it. /i “The government needs to wean itself off its bad habit of believing every dodgy claim it’s told by Big Tech – including OpenAI – starting with the idea they can cover the UK in power-guz
843.
by chrisjj 7mo ago
I'd donate to have Mozilla raise Thunderbird to second class.
844.
by chrisjj 7mo ago
That's exactly what it means.
845.
by chrisjj 7mo ago
Clickbait.
846.
by chrisjj 7mo ago
> With misleading titles and non-existent metadata, the data currently available cannot support any meaningful analysis No problem. Their chatbot "AI" will fabricate non-existent meaning and misleading analysis. That's wha
847.
by chrisjj 7mo ago
And worse look if the arguer contradicting himself is the CEO. But then, he knows his audience. The pre-gulled.
848.
by chrisjj 7mo ago
> in OpenAI is that everything we could imagine works now. Lack of imagination will do that. Like, imagine "works" meant fit for purpose.
849.
by chrisjj 7mo ago
> Iran has no leverage. Patently false. Else there'd be no ceasefire.
850.
by chrisjj 7mo ago
> safeguards Ha ha.
851.
by chrisjj 7mo ago
Fortunately reverse engineering too is in the dictionary - to help anyone mistaking it for spec generation.
852.
by chrisjj 7mo ago
> Yes. These HN guidlines already basically cover it: >> Please don't post shallow dismissals, especially of other people's work. A good critical comment teaches us something. >> Please don't complain about tan
853.
by chrisjj 7mo ago
Also add "AI can make mistakes". Thank you.
854.
by chrisjj 7mo ago
You're suggesting this is the complainant's fault?
855.
by chrisjj 7mo ago
> distilling A.k.a. as fabricating. No wonder they chose to use "AI".
856.
by chrisjj 7mo ago
> The specs were derived from the code itself Oh dear. I strongly suggest this author look specification up in a dictionary.
857.
by chrisjj 7mo ago
I doubt Anthropic see this as their capacity problem. They like "extra usage", and users who don't, well its their capacity problem.
858.
by chrisjj 7mo ago
> elevated errors So, we have errors floating in the air? Or this is the output of an "AI"?
859.
by chrisjj 7mo ago
What, was intelligent ?
860.
by chrisjj 7mo ago
Has anyone tried turning Microsoft off and on again?
861.
by chrisjj 7mo ago
> So with both the US and Israel's own arrogance World-beaters, both of them.
862.
by chrisjj 7mo ago
Trump to Iran: Cave by Tuesday, else we'll obliterate your petrochem. Israel obliterates petrochem. Trump to Iran: Cave by Tuesday or... or...
863.
by chrisjj 7mo ago
> Naming things is hard ... for people really bad at it.
864.
by chrisjj 7mo ago
> This account holder has a history over several years of registering new accounts How do you know this is the registrant of all?
865.
by chrisjj 7mo ago
They will... if they ever do ship.
866.
by chrisjj 7mo ago
> I don’t understand what he’s actually building with 37k loc/day. ... I tried to go look at his blog but it didn’t load on my phone. That 37k loc/day must be JS in his blog. /i
867.
by chrisjj 7mo ago
> states generally agree that giving weapons the capacity to select and engage targets on their own could lead to unintended consequences, including serious harms for which no human would be held accountable. And why ever would that huma
868.
by chrisjj 7mo ago
Pharmacy. Not pharmacist. Pharmacies provide much more than just medicine.
869.
by chrisjj 7mo ago
> We built a self-correcting extraction system that went from 80% to 95% row count accuracy Got to wonder who has any use for 95% accuracy - at just counting rows .
870.
by chrisjj 7mo ago
And so it begins...

More ›