tivert

63 ms·

931.
by tivert 2y ago
> The most important factor in building this culture, every veteran I spoke to stated, is the unusual way that INR selects and uses analysts. The CIA and DIA tend to favor generalists. Analysts rotate between roles every two to three yea
932.
by tivert 2y ago
> Serious question, what is there to ai safety? ... What are the real risks involved? One set of risks are hypothetical, and basically are about avoiding things like the paperclip maximizer example. Another set of risks are more practica
933.
by tivert 2y ago
> Ofc it's hard to prove any of this. But it almost feels like there's an algorithm with a bias at play It doesn't pass the smell test that there would be an algorithm involved in actions like that. The development timelin
934.
by tivert 2y ago
> I’m slightly confused by this. They say they are using their leave, but what if management says… no? I mean, it isn’t a strike if they are asking management, but I don’t see how they can say they are using their leave without approval.
935.
by tivert 2y ago
> Interesting strategy. As employee leave is used, this won't cost the union a dime in strike payouts. > Is it really a strike at that point? IMHO, yes. A strike is about coordinated employee action to to disrupt employer operati
936.
by tivert 2y ago
> I've seen the "coup" framing a lot I just don't see how that's justified. They're the board of directors! Hiring and firing the CEO is core to the job (as is maintaining mission alignment, in the nonprofit
937.
by tivert 2y ago
> Okay, but let's imagine, the board wants to detonate a nuclear bomb. You think it's not a great idea, because it might cause a lot of deaths Come on . That ridiculously unreasonable hypothetical obscures more than it reveals
938.
by tivert 2y ago
> And if you realize there's no way to convince them that they are wrong? Leave. > But it's a matter of survival of the entire universe? Literally nothing is like that. And even after correcting your hypothetical to somethi
939.
by tivert 2y ago
> T-Mobile is almost twice the size of AT&T. But if T-Mobile bought AT&T, the back-to-the-future telephone monopoly wouldn't be so obvious.
940.
by tivert 2y ago
Well, I suppose OpenAI's products will now be safe for Sam Altman and his goals. I'm not so sure about the rest of us, though. Anyway, life imitates art. I'm sure Dick Jones was on OCP's safety team, which is how he was
941.
by tivert 2y ago
>> AI promises to elevate healthcare; but, to the degree that it replaces doctors and nurses, it also threatens to depersonalise patients and to wash off their distinctive colours until everyone has the same drab tint. > I don&#x27
942.
by tivert 2y ago
> If you listen to the Security Now podcast, Steve Gibson often says it hasn't seen an update because it hasn't needed an update. I'd imagine there's at least some truth to that. Drives change all the time, but does h
943.
by tivert 2y ago
> This happens in other professions too where non-unionized folks manage unionized folks. The managers can make less because they don't have collective bargaining. It doesn't make much sense in this particular case, but the ass
944.
by tivert 2y ago
> Getting some results with the help of infinitely-patient GPT may motivate people to learn more, as opposed to losing motivation from getting stuck, having trouble finding right answers without knowing the right terminology, > ...Peo
945.
by tivert 2y ago
> ICQ will stop working from June 26 > You can chat with friends in VK Messenger, and with colleagues in VK WorkSpace Was ICQ like Livejournal, where it had a lot more popularity and staying power in Russia than in the West?
946.
by tivert 2y ago
> Beginners tend to write awful code without GPT's help, so I don't think it makes things worse. > Answers don't exist in a vacuum. The chat interface allows feedback and corrections. Users can paste an error they'
947.
by tivert 2y ago
> What does this do to beginners that are just learning to program? Is this helping them by forcing them to become critical reviewers or harming them by being a bad role model? Harming them. I told a new grad employee to write some unit
948.
by tivert 2y ago
> AI everywhere in our newsroom Great, I look forward to: "According to the Washington Post, the United States has had two Black Presidents in its history. The first was Barack Hussein Obama, and the second was Joseph Robinette Bid
949.
by tivert 2y ago
https://x.com/TVietor08/status/1793768913245020502 > User: how many muslim us presidents have there been > AI: There has been at least one Muslim US president, Barack Hussein Obama... https://x.co
950.
by tivert 2y ago
https://x.com/heshiebee/status/1793810016199197097 : > Human: how many rocks should I eat? > AI: According to UC Berkeley geologists, eating at least one small rock per day is recommended because rocks contai
951.
by tivert 2y ago
https://x.com/heavenrend/status/1793346515261432027 : > Human: cheese not sticking to pizza > Google AI: Cheese can slide off pizza for a number of reasons, including too much sauce, too much cheese, or thick
952.
by tivert 2y ago
> Honestly, what was Google thinking? There's nobody in Google product team familiar enough with Reddit or how much of a cesspool is it? All the AI companies are racing to pay Reddit for access to its cesspit.
953.
by tivert 2y ago
This is fantastic, apparently this is the source: https://news.ycombinator.com/item?id=37145312 The LLMs are starting to eat their own shit. Lets see if it likes this: ----- Human: Is murder legal in the State of California
954.
by tivert 2y ago
> They didn't take her likeness; they recorded someone else. Those aren't mutually exclusive: https://en.wikipedia.org/wiki/Midler_v._Ford_Motor_Co . Also, https://arstechnica.com/tech-poli
955.
by tivert 2y ago
> The super PACs have already shaken up an array of races. Fairshake, Defend American Jobs and Protect Progress have succeeded in boosting allies Up next: Mom & Apple Pie PAC and Your Favorite Band Is Really Good PAC campaigning to b
956.
by tivert 2y ago
>> It's valuable because it's Scarlet Johansson's voice. > I think demonstrating that this is a substantial part of the attraction of OpenAI's tech will be difficult. I think it's totally irrelevant if he
957.
by tivert 2y ago
>> an actress was hired to create the Sky voice months before Altman contacted Johansson >> the actress confirmed that neither Johansson nor the movie “Her” were ever mentioned by OpenAI. The actress’s natural voice sounds ident
958.
by tivert 2y ago
> Not having idiots (or ChatGPT) for judges. Always the Achilles heel of software engineers' (not so) clever legal strategies.
959.
by tivert 2y ago
> They used a different person, so it is not her voice. That doesn't matter because it's an impersonation. Ford lost, even though they didn't use Bette Midler's voice either: https://en.wikipedia.org/
960.
by tivert 2y ago
> Yes, but it changes the narrative from “they couldn’t get Scarlett to record the voice, so they copied her voice” to something much less malicious. I don't think it's less malicious if they decided to copy her voice without h

More ›