Gregory's Server
|
Show previous comments
@carnage4life sad to see that the idea of Americans have given up the fight for a sane public healthcare policy so the assumption is their grandkids will suffer from this.  Welcome to the cycle of generative AI making search worse. Quora uses ChatGPT which hallucinates an answer to a nonsense question. Google Search picks up this nonsense answer from Quora which has high page rank and treats it as an instant answer.
Show previous comments
 The Saudi Arabian government sentences a man to death for his tweets and YouTube. Given Saudi Arabia is a major investor in Twitter/X, I wonder if we’ll hear from Elon Musk or Linda Yaccarino about how one of their users is being sentenced to death for exercising their free speech rights on their platform by their second largest investor. https://apnews.com/article/saudi-arabia-death-sentence-twitter-a2b5549806605d1d21f332ac4c36e43f
Show previous comments
@carnage4life I wish the future The-Line-Inhabitants a lot of fun in their "paradise" 😤 Streaming is now just as expensive and confusing as cable. Ubers cost as much as taxis. And the cloud is no longer cheap. In other words, enshitification comes for us all. 😏
Show previous comments
@carnage4life ride hailing was only cheaper because no wage protection for drivers & no control over quality of service or vehicle. As standards are put in place the costs start matching up.
Show previous comments
 @carnage4life I always omit preconditions from examples I use to illustrate a specific point so that if they're reused verbatim they will crash. I expect people to actually read and understand what they're doing. (Obviously the preconditions are spelled out earlier, but later examples only show the code directly relevant to the point being explained.) @carnage4life many years ago I was rolling out SendGrid using their .net library. We very quickly ran out of connections because the actual library wasn't doing connection management. I filed the bug, but they pointed at the official.net example for using the http client. This was the newest HttpClient that was actually a factory for The funny thing is that these price increases don’t take into account costs going up after they settle the SAG & WGA strikes. So more price increases are likely coming which means more people questioning if they should be paying for all these services. It’s definitely a transitional time for the streaming industry.
Show previous comments
 So you're saying the CEO/investor class behind these increases won't scale back any of their profit-taking...   @carnage4life I thought Disney+ was much older but then I realized I might be thinking of the Disney channel (1983).
Show previous comments
  A broken clock fixed by taping a working clock over it is a metaphor for every codebase you’ll encounter in your professional career as a software developer.
Show previous comments
Looks to me like the boss or manager went and bought another boat with the discretionary fund again. @carnage4life And along comes a business administrator and orders you to take the taped clock off. Don't you know the clock underneath goes right twice a day? The taped one only once. Can't have that. @carnage4life The ceiling clock is very thick so looks as though it is double faced, which I have seen in very large rooms. So this could be a bodge where the original clock works from one viewpoint but requires a hack from another viewpoint.
Show previous comments
 @carnage4life “You can run your company successfully with Zoom. We can’t of course. But we’re sure you can.” As long as generative AI is defined by “fancy autocomplete” also known as large language models, we should expect that hallucinations won’t go away. The idea of forcing autocomplete to tell the truth is absurd but that is essentially what people are expecting out of LLMs like ChatGPT and Google Bard. It’s quite likely these are simply impossible expectations to meet. https://fortune.com/2023/08/01/can-ai-chatgpt-hallucinations-be-fixed-experts-doubt-altman-openai/
Show previous comments
@carnage4life People don’t get this. Accuracy of the information is entirely unrelated to the goal of these models, which is simply to sound human. @carnage4life Asimov in his legendary series that began with I, Robot and ended with the last foundation book, explores this in exhausting detail. An adversary eventually breaks any rules-based system. Asimov makes the observation only by eliminating the value of personhood and creating a hive-ish mind can you create a rule system. LLMs are a more sophisticated Positronic brain than a random number generator, but they can still not think beyond their rule limits.  @carnage4life "Autocomplete" is a good way of putting it, based on parroting all things that have ever been said on the internet. LLMs: the world's most sophisticated autocomplete parrots. Feeling so much better about my healthcare already. Forrester Research published a research note with this key takeaway after Twitter rebranded to X. Prediction: X Will Shutter Or Be Acquired Within The Next 12 Months 🙀 https://www.forrester.com/blogs/for-most-twitter-users-x-wont-mark-the-spot/
Show previous comments
@carnage4life must be nice to burn cash so easily. Hey, do you know anyone to help get verified on threads?  Forrester: “While Musk’s vision is to turn “X” into an “everything app,” this takes time, money, and people — three things that the company no longer has.” Someone just asked what revolutionary product has Google shipped outside of search. At first I was surprised at the question then realized that it’s been over ten years since Google any meaningful new product. They’ve just been canceling projects and shipping redundant chat apps for a decade.  @carnage4life I want to point out. Google Sound Search and Google Lens are amazing. They simply don't have comparable alternatives and are often overlooked. It also misses Chrome which has seen 10 years of constant innovation and changed the way we use the Internet (it probably wasn't axed as hard as other products because Pichai himself comes from the Chrome team). I give EU regulators a hard time for being too conservative in tech regulation but I’ll always give them credit for at least understanding the problem space. The real AI risk is discriminatory algorithms not Skynet becoming self aware and dropping nukes. US regulators on the other hand are either mouthpieces of lobbyists or regurgitating viral tweets  @carnage4life ...& of course OpenAI's response to this is to say 'if you persist in fighting discrimination we can't do business in the EU.'  @carnage4life absolutely spot on. All the “existential risk” noise from the folks investing billions in it are attempts to put a regulatory moat around what they cannot put a technical moat around People will look back and realize OpenAi had the competitive advantage of crawling the web to train their AI before publishers realized what was going on. Reddit & Twitter raising their API prices to tens or hundreds of thousands for basic access is just the beginning. Conversations I’ve had in recent weeks and data from articles like this point to a new tech reality — FAAMNG has stopped hiring but everyone else in tech still is. Unemployment in tech is at a tiny 2.2%. The laid off are finding jobs in smaller companies One person I spoke to said their RSUs are so underwater that the best financial outcome would be getting laid off to collect severance and then getting a gig at a smaller company to address the compensation shortfall from stock price collapse from 2021.  @carnage4life there is a difference though. I can see if the code I am about to copy is crap. ChatGPT just cheerfully spew it out without batting an eye. Nintendo’s CEO recently said “If we reduce the number of employees for better short-term financial results, employee morale will decrease” and is instead doing the opposite by raising pay to address inflation. Some leaders are built different. https://www.gamespot.com/articles/nintendo-will-pay-its-workers-10-more/1100-6511268/
Show previous comments
@carnage4life Can’t imagine an American or European leader doing this: “calls by Prime Minister Fumio Kishida for Japanese companies to pay workers more as inflation takes hold in an economy used to years of deflation and stagnant wages, and as Japan prepares for its annual spring round of labour negotiations” https://www.reuters.com/technology/nintendo-trims-annual-profit-outlook-firmer-yen-2023-02-07/?rpc=401 |
@carnage4life This is part of Steve Bannon's "flood the zone with shit" strategy... Not necessarily to make people believe the lie (though that's a bonus), but to make people distrust the truth and give up caring.
@carnage4life another reason to dispel the efficacy of "AI detectors".
@carnage4life ten 90% accurate fake detectors and a motivated reasoner...
edit: also it's a black box. The machine learning could be investigated, probably,(tho difficult..) but not if it's running on someone-elses server..