Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
lisasays
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
61 ms
·
811.
▲
by
lisasays
4y ago
Southwest Airlines, just this past winter. Multiple factors were involved, but chief among these appears to have been a series of deficiencies in its "outdated" scheduling system (which upper management was well aware of, but it
812.
▲
by
lisasays
4y ago
You can say the piece was biased if you want. But to apply the description "cold-blooded" seems, well, quite weird.
813.
▲
by
lisasays
4y ago
It refers the type of advertising that intentionally includes contradictory messaging (effectively, talking out of both sides of its mouth). That's why it isn't "wild" at all for them to intentionally put out messaging t
814.
▲
by
lisasays
4y ago
When people cry that there will be not enough work, the exact opposite happens. Nevermind that these people aren't crybabies - and that it is incredible condescending to paint them as such. More fundamentally - what you're asse
815.
▲
by
lisasays
4y ago
That's not the product that's doing X, it's the company I would reckon that to the good (if not the vast) majority of those who will see their jobs either yanked away from them entirely, or otherwise negatively perturbed in
816.
▲
by
lisasays
4y ago
Not wild at all. It's a standard cognitive dissonance technique, widely used in advertising.
817.
▲
by
lisasays
4y ago
"There is no person behind what the algorithm emits, therefore no intent - hence, it's not an accusation." At least that's how I'm imagining the line of reasoning behind this point of view goes.
818.
▲
by
lisasays
4y ago
https://archive.is/mn841 A most curious collection indeed.
819.
▲
by
lisasays
4y ago
https://archive.is/JXxWx
820.
▲
Gochujang Is So Much More Than a Condiment
(nytimes.com)
1 points
by
lisasays
4y ago
|
1 comments
821.
▲
by
lisasays
4y ago
No, it's OpenAI saying "look how credible, authoritative and reliable our product is -- see, even Morgan Stanley uses it."
822.
▲
by
lisasays
4y ago
You might want to look at it this way: As an analyst, your job isn't to query the data. It's to understand the data - and as the other commenter put it, to know what questions to ask.
823.
▲
by
lisasays
4y ago
It's always been about the dangers of humans - specifically those who create and market products of this nature. With us as their unwilling test subjects.
824.
▲
by
lisasays
4y ago
Yet another perfect application for GPT.
825.
▲
by
lisasays
4y ago
As with tobacco companies that of course have warning labels in all the legally mandated places -- yet still run zillions of ads showing what a youthful, vibrant, carefree lifestyle their product enables -- it's plain to see the real m
826.
▲
Emahoy Tsegué-Maryam Guèbrou, Nun with a Musical Gift, Dies at 99
(nytimes.com)
4 points
by
lisasays
4y ago
|
0 comments
827.
▲
by
lisasays
4y ago
Who is selling anything as a reliable source of information? OpenAI for example, whose marketing materials present it as exactly that. For example: You essentially have the knowledge of the most knowledgeable person in Wealth Management
828.
▲
Why does written language have to be so particular?
(nytimes.com)
2 points
by
lisasays
4y ago
|
0 comments
829.
▲
ChatGPT is making up fake Guardian articles. Here’s how we’re responding
(theguardian.com)
9 points
by
lisasays
4y ago
|
1 comments
830.
▲
by
lisasays
4y ago
I'm not sure being a Jira monkey (uh, "engineer") is either -- but more importantly, we don't know what this person's role is.
831.
▲
by
lisasays
4y ago
In fact they have a bunch of disclaimers warning users the exact opposite. Maybe buried in the fine print somewhere. Its own splash page, however, paints much more of a "what's not to like?" picture: We’ve trained a model
832.
▲
by
lisasays
4y ago
Let's assume it's true: a Paytm employee acted negligently. Not negligently - maliciously. The employee knew exactly what they're doing, that it was "wrong" in any conventional sense -- and most likely a huge liab
833.
▲
by
lisasays
4y ago
No, just that I don't run any of these surveillance apps unless I absolutely have to. And hence, am unfamiliar with the vast majority of them.
834.
▲
by
lisasays
4y ago
[dupe]
835.
▲
by
lisasays
4y ago
[dupe] - https://news.ycombinator.com/item?id=35471826
836.
▲
King Charles signals first support for research into monarchy’s slavery ties
(theguardian.com)
4 points
by
lisasays
4y ago
|
0 comments
837.
▲
by
lisasays
4y ago
I'm so glad I've never heard of that service and have no idea what it is. Thanks very much for calling them out by name, BTW. Presumably someone from that company is reading this as we speak - and soon enough, will be reporting b
838.
▲
by
lisasays
4y ago
Are you really victim blaming here? Nasty. Sure sounds like it.
839.
▲
by
lisasays
4y ago
The creators liable for a hallucination? For hallucinations marketed and sold as a reliable source of information - yes. Consequences of what action? For what they do to others on the basis of the above.
840.
▲
by
lisasays
4y ago
The nature of Holodomor is only disputed by the historians who care to push a specific agenda. You can believe that if you want. Without factual references to back up what you're saying, though - it's just a a flat, sweeping as
More ›