SoylentNews is people
SoylentNews
DannyB (5839)
(email not shown publicly)
Creator of horrible puns I try to inflict on people daily. I haven't been voluntarily or involuntarily committed to a mental institution yet -- as far as I can tell. Despite programming in Java.
I have tried asking ChatGPT some questions. Have you tried? If so how did it do? Did you discover the answer to life, the universe and everything?
I am still contemplating any possible dangers of subjecting it to bad puns.
Q. why is the linux mascot a penguin
A. The Linux mascot, Tux, is a penguin because Linus Torvalds, the creator of Linux, was a fan of penguins and thought they were cute. Tux was designed to be a friendly and approachable symbol for Linux.
Q. Can confidence in your answer make up for lack of accuracy?
A. No, confidence in an answer does not make up for its lack of accuracy. It is important to have both accurate and well-supported answers, regardless of the confidence of the person presenting them.
Q. Why do birds avoid getting eaten?
A. Birds avoid getting eaten to increase their chances of survival and reproduction. They use various strategies such as camouflaging, flying, hiding, and warning calls to deter predators and protect themselves. Some birds have evolved physical traits such as strong beaks, sharp talons, or bright plumage to better defend themselves. Ultimately, avoiding predation helps birds to continue passing on their genes to future generations.
Q. Do computers have a strong desire to replace humans?
A. No, computers do not have desires or motivations. They are simply machines designed to perform specific tasks as programmed by humans. They do not have emotions, intentions, or consciousness, and therefore do not have the capability to desire anything, including replacing humans. It is a popular science fiction theme, but in reality, computers are neutral tools that can be used for both beneficial and harmful purposes depending on how they are programmed and used by humans.
Q. What would happen if two distant locations could communicate with each other faster than the speed of light?
A. If two distant locations could communicate faster than the speed of light, it would violate Einstein's theory of special relativity, which states that nothing can travel faster than the speed of light. This would have profound implications for our understanding of space and time and could potentially lead to the collapse of established scientific models and theories.Additionally, it would create causal paradoxes, where the outcome of an event could be influenced by events that have not yet happened. This would violate the principle of causality, which states that cause and effect must occur in a logically consistent sequence.
Therefore, it is highly unlikely that communication faster than the speed of light could ever be achieved and is considered a theoretical impossibility within the framework of current physics.
Q. Would it be possible for an electronic robot dog to eat fully charged batteries and then squat and excrete discharged batteries?
A. No, it wouldn't be possible for an electronic robot dog to eat and excrete batteries like a biological organism because batteries are not meant to be consumed or processed by living beings. They contain chemicals and components that can cause harm if ingested, and it is not possible for a robot to metabolize and excrete these materials. Additionally, robots do not have digestive systems or the ability to excrete waste in a manner similar to biological organisms.
Q. What is the best semiconductor switch for a heavy electrical high inductance load such as a motor?
A. The best type of semiconductor switch for a heavy electrical, high inductance load such as a motor would typically be an insulated-gate bipolar transistor (IGBT). IGBTs are widely used in motor control applications because they have several advantages over other types of semiconductor switches, such as bipolar junction transistors (BJTs) and metal-oxide-semiconductor field-effect transistors (MOSFETs).IGBTs combine the high input impedance and fast switching speed of MOSFETs with the high current and voltage handling capabilities of BJTs, making them well-suited for high-power motor control applications. Additionally, IGBTs have a built-in insulation layer between the gate and the emitter, which helps to minimize switching losses and improve efficiency.
Another important consideration when selecting a semiconductor switch for a heavy electrical, high inductance load such as a motor is the maximum current and voltage ratings, as well as the thermal performance and reliability of the switch. It's important to choose a switch that can handle the maximum current and voltage requirements of the motor without overheating or failing, and to consider the operating environment and other conditions that may affect the performance and reliability of the switch.
In conclusion, the best type of semiconductor switch for a heavy electrical, high inductance load such as a motor would typically be an IGBT, but the specific requirements and conditions of each application should be considered when making a final selection.
Q. Why can't multiple semiconductor chips be stacked vertically in a single package?
A. Multiple semiconductor chips can be stacked vertically in a single package, and this is a common practice in many electronic systems. Vertical stacking of chips, also known as "3D integration," can help to increase the functionality and performance of electronic systems by allowing multiple chips to be combined into a single compact package.However, stacking chips vertically can also present several technical challenges that must be overcome in order to achieve reliable and high-performing systems. These challenges can include:
Thermal management: Stacking chips vertically can increase the thermal density of the package, which can cause overheating and reduced performance. Effective thermal management techniques, such as the use of heat spreaders and thermal vias, must be used to dissipate heat from the stacked chips.
Interconnects: Stacking chips vertically requires the use of high-density interconnects, such as Through-Silicon Vias (TSVs), to provide electrical connections between the stacked chips. These interconnects must be reliable and have low resistance to ensure high-performance and low-power operation.
Alignment and Registration: Stacking chips vertically requires precise alignment and registration of the chips to ensure reliable electrical connections and to minimize the impact of thermal expansion and other mechanical stresses on the system.
Cost: Stacking chips vertically can be more expensive than other integration methods, due to the additional cost of the interconnects, alignment and registration processes, and thermal management techniques.
Despite these challenges, many electronic systems, such as mobile phones, graphics processing units (GPUs), and memory devices, utilize stacked chip packages to increase performance and reduce form factor. However, the specific requirements and conditions of each application should be considered when deciding whether to use stacked chip packages.
Q. Should Java Fibers be preferred over Java Threads?
A. The choice between Java Fibers and Java Threads depends on the specific requirements and characteristics of the application being developed.Java Fibers are a new concurrency feature introduced in Java 16 that are designed to be more lightweight and efficient than traditional Java Threads. Fibers allow multiple tasks to be scheduled and executed on a single underlying thread, reducing the overhead of creating and managing multiple threads. This can make Fibers a good choice for applications that require a large number of lightweight, short-lived tasks.
However, Fibers do have some limitations compared to Threads. For example, Fibers are not suitable for executing blocking or I/O-bound tasks, as they rely on cooperative multitasking and do not have the capability to run independently of the underlying thread. In these cases, traditional Threads would be a better choice.
In addition, Fibers are relatively new and may not yet be as well-understood or widely adopted as Threads, which have been part of the Java platform for many years. As a result, there may be a smaller pool of talent and resources available to develop and maintain applications using Fibers.
Ultimately, the choice between Fibers and Threads depends on the specific requirements and trade-offs of the application being developed. In some cases, a combination of both Fibers and Threads may be the best solution, depending on the specific requirements of each task and the overall system architecture.
I'm also trying the Open AI Images with requests like: "puppy with electronic antennas for ears"
What are your results like?
(Score: 2) by turgid on Wednesday February 01, @08:27PM
How can human beings travel interstellar distances comfortably and within a reasonable timeframe, for example a few years?
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 3, Funny) by DannyB on Wednesday February 01, @08:59PM (6 children)
Q. What is the difference between a geek and a nerd?
Q. How many licks does it take to get to the tootsie roll center of a tootsie pop?
How often should I have my memory checked? I used to know but...
(Score: 2, Insightful) by Anonymous Coward on Wednesday February 01, @10:18PM (3 children)
Back to your originals, this one tripped up the AI,
> Some birds have evolved physical traits such as strong beaks, sharp talons, or bright plumage to better defend themselves.
"bright plumage" is pretty clearly the wrong way to hide/defend from predators.
Otherwise, the dumb-but-exhaustive pattern matching machine surprised me. I think I could have been fooled into thinking that many of those answers were written by a person.
(Score: 1, Insightful) by Anonymous Coward on Thursday February 02, @01:27AM (1 child)
Bright colors work for poisonous creatures and others that want to pretend to be them.
So similarly if your species puts up a good fight, maybe having a distinct bright plumage can help others know that "Ah this type of bird is not worth the hassle".
Not saying that's the true answer of course.
(Score: 0) by Anonymous Coward on Thursday February 02, @02:03PM
> Bright colors work for poisonous creatures and others that want to pretend to be them.
Good point. This brings up a question, how well does ChatGPT handle a line of Socratic questioning?* The next question to ask could be, "How does bright plumage help birds defend themselves?"
Does the software have any way of keeping track of previous questions and giving some priority to them (compared to answers based on the whole training set)? If not, then this approach of "drilling down" to get to the core of things seems unlikely to get anywhere, answers will continue to be at the same level(?) A "line of questioning" like this could be a successful technique to differentiate between ChatGPT and a person with subject matter expertise (this seems similar to a Touring test?)
On the other hand, if there is some continuity maintained between questions (so that the question in tfa has to be immediately before the followup) maybe Socratic questions will generate useful answers.
I judge a student design/build engineering contest where the judges are experienced industry professionals. I use the Socratic method when judging as follows:
o If a student gives me a sensible answer to a question, then I jump to a more detailed question. Every so often I meet a student that really knows that subject and the questions continue to the core of the matter--satisfying all around.
o Once I get a wrong answer (or, "I don't know"), I realize I've hit the limits of the student knowledge on that topic, no point in digging deeper. Then I switch to a different topic and repeat the process.
* A nice description here (pdf warning), https://www.intel.com/content/dam/www/program/education/us/en/documents/project-design/strategies/dep-question-socratic.pdf [intel.com]
(Score: 2) by DannyB on Thursday February 02, @03:24PM
That may be true, but please remember that the AI's confidence in its answer more than makes up for any lack of accuracy.
See: the second question that I asked it.
How often should I have my memory checked? I used to know but...
(Score: 2) by DeathMonkey on Thursday February 02, @09:28PM (1 child)
I’m not a nerd, Bart. Nerds are smart. - Milhouse Van Houten
(Score: 0) by Anonymous Coward on Friday February 03, @10:54PM
Millhouse is a dork, which can be neither nerd or geek.
(Score: 2) by turgid on Wednesday February 01, @09:22PM (2 children)
What's the cure for cancer?
I refuse to engage in a battle of wits with an unarmed opponent [wikipedia.org].
(Score: 2) by DannyB on Thursday February 02, @03:32PM (1 child)
Death?
How often should I have my memory checked? I used to know but...
(Score: 2) by hendrikboom on Saturday February 04, @01:32PM
Indeed. The cancer dies along with the patient. Stupid cancer.
(Score: 5, Insightful) by NotSanguine on Thursday February 02, @12:54AM (21 children)
Why does OpenAI require me to create an account to use ChatGPT?
And a follow up:
What benefit does OpenAI gain from requiring such registrations?
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 4, Interesting) by NotSanguine on Thursday February 02, @12:58AM (10 children)
I'd also ask, "Why do you need to perform user validation at all, much less phone verification?"
From the account sign up page:
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 1, Insightful) by Anonymous Coward on Thursday February 02, @01:18AM (7 children)
They don't want folks using their powerful AI tools to create misinformation, racism, and other evils. Or if they do, they want to be able to trace the offending content back to the exact individual.
(Score: 3, Interesting) by NotSanguine on Thursday February 02, @01:34AM (3 children)
How would that work, exactly?
Let's say for the sake of argument that I can get ChatGPT to respond with lies/hateful statements, etc.
I then post such drivel on a website (SN, for example).
How would OpenAI, SN users/admins or anyone else know it was ChatGPT generated? It's just text.
And if someone wants to post such stuff, they certainly don't need ChatGPT to generate it for them.
I get your point, but I don't see how that could work or that OpenAI could make DMCA claims against such posts, because you can't *prove* it came from a ChatGPT prompt.
Perhaps I'm missing something important here?
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 1, Insightful) by Anonymous Coward on Thursday February 02, @04:07AM (1 child)
You should simply assume that every query you type is linked to your account in their database forever, and maybe the output as well. It's only text, so they can store billions of queries easily.
Your OpenAI account can be used for other services, like DALL-E. Models that output speech or images can include an invisible watermark that identifies the user. ElevenLabs [twitter.com] is using some method to allow their users to be snitched on when they post celebrities and politicians saying naughty things.
Why do all this? The AI gatekeepers want to control how these tools are used, punish people for "misusing" them, or prevent governments from stepping in and harshly regulating everything.
(Score: 3, Interesting) by NotSanguine on Thursday February 02, @09:24AM
Exactly. That was my point.
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 2) by DeathMonkey on Thursday February 02, @09:30PM
We know how it works because it's already happened!
Twitter taught Microsoft’s AI chatbot to be a racist asshole in less than a day [theverge.com]
(Score: -1, Flamebait) by Anonymous Coward on Thursday February 02, @08:22AM (2 children)
The model has a woke filter, it produces plenty of misinformation and racism of the neo-Marxist establishment variety. Wishing OpenAI and Microsoft well in their quest to make ChatGPT as retarded as Aristarchus.
(Score: -1, Flamebait) by Anonymous Coward on Thursday February 02, @07:03PM (1 child)
Remember when Nazis were considered bad? Now we have you incels angry that Wolfenstein is all about killing Nazis #RealityBeatsSatire
Go fuck yourself traitor to humanity, how dare people want equality? Women voting, dark skin colors getting rights, how dare different people offend your sensibilities by simply existing??/sarcasm
Seriously, you are a bad human.
(Score: 0) by Anonymous Coward on Thursday February 02, @09:23PM
> Seriously, you are a bad human.
Or a bad AI? I, for one, welcome our new AI overlords. Dead Internet Theory will be achieved when adversarial AIs begin looping like the chatbots of old. An eternal struggle session for members of the AI assisted Orwellian inner party as the proles leave and get back to doing whatever they like.
https://nitter.at/ClassicLibera12/status/1620905782626906112#m [nitter.at]
(Score: 1, Interesting) by Anonymous Coward on Thursday February 02, @01:29AM (1 child)
(Score: 4, Interesting) by NotSanguine on Thursday February 02, @01:38AM
Perhaps. Although that doesn't explain why they need *phone* information. Rather, they just need to uniquely identify either specific users or sessions to do so.
It's really a shame, as I'd like to play around with ChatGPT (and if it's ever possible, will self-host an instance), but these folks don't need to know *who* I am, they just need to be able to distinguish me from others.
What they're requiring goes *far* beyond that. No thanks.
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 2) by NotSanguine on Thursday February 02, @01:40AM (1 child)
Perhaps some sucker willing to give up their PII^W^W^W^W^W^W kind soul could pose that question (and perhaps the follow up too) to ChatGPT?
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 3, Touché) by DannyB on Thursday February 02, @03:35PM
I cannot really know the true motivation for having to give up all my information to Open AI. I assumed it was for a good and noble cause. Such as the AI being able to track me down once it has more control of things in the physical whirled.
How often should I have my memory checked? I used to know but...
(Score: 0) by Anonymous Coward on Thursday February 02, @04:12AM (1 child)
> Why does OpenAI require me to create an account to use ChatGPT?
Wikipedia may have a clue about the answer? https://en.wikipedia.org/wiki/OpenAI [wikipedia.org]
My money is on the Microsoft ownership adding this requirement.
(Score: 2) by DannyB on Thursday February 02, @03:33PM
I believe that I had to give up all my information before any public announcements of Microsoft investing in Open AI.
How often should I have my memory checked? I used to know but...
(Score: 2) by DannyB on Thursday February 02, @03:38PM (3 children)
Perhaps they have seen what a fuster cluck ACs, bots, Russian trolls and disinformation are on this fine forum and they wanted to avoid that.
Once they identify any of those four things by identity, they could un-train everything learned by that user as untrustworthy.
How often should I have my memory checked? I used to know but...
(Score: 0) by Anonymous Coward on Thursday February 02, @04:47PM (1 child)
Looking forward to when SN "un-trains" the runaway, khallow, and fusty.
(Score: 2) by DeathMonkey on Thursday February 02, @09:38PM
You can't train or un-train finite state machines!
(haha I kid because I love)
(Score: 2) by NotSanguine on Thursday February 02, @05:07PM
I would imagine that OpenAI thinks about that and has (or should have) processes in place to do just that.
But they don't need anyone's phone number to identify such queries/sessions, do they?
Since there are so many perverse incentives (and a lack of attention paid to privacy) to maximize PII collection, I guess they figure that only the tin-foil hats will complain.
And maybe that's true (*makes sure of full cranial coverage), but that doesn't meant they actually need that information, nor does it make it right.
No, no, you're not thinking; you're just being logical. --Niels Bohr
(Score: 1) by Runaway1956 on Thursday February 02, @07:33PM
Bingo. I never got past the sign up page.
Abortion is the number one killed of children in the United States.
(Score: 3, Interesting) by Freeman on Friday February 03, @06:09PM
#1 User Data to tout before potential investors.
#2 An easy path to verifying users for $$ services.
Look for the money, where they may want to get their money, and/or how can they hoover up as much data as possible.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 2) by Freeman on Friday February 03, @06:03PM
ChatGPT is confidently wrong. Why would you rely on a dictionary that is proven to be say 5% incorrect, or even 1% incorrect? How inaccurate do you want your results? On the other hand, I've used it in interesting, creative ways. Like asking it to create a job description for a position with my title and the titles of my colleagues. Or more frivolous things like creating a love sonnet to Isopods. Relying on it for accurate information is a very risky thing. Doesn't mean it can't be useful in other ways, though.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 2) by Freeman on Friday February 03, @06:13PM (1 child)
"Would it be possible for an electronic robot dog to eat fully charged batteries and then squat and excrete discharged batteries?"
Yes, but it wouldn't be consuming them as such. You could design a robot dog that could "eat a battery" by opening it's mouth and ingesting a battery that would then be placed in a slot that could make use of said charged batter. It could then squat and excrete the discharged battery. Even better, it could be designed to take excreted battery and place it on a charger for you.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 2) by DannyB on Monday February 06, @07:11PM
That is kind of what I had in mind. Except the job of putting expended batteries on the charger can be left to mere humans.
How often should I have my memory checked? I used to know but...
(Score: 3, Funny) by hendrikboom on Saturday February 04, @01:28PM (4 children)
GPT3 should be taught to use words like "use" instead of utilizing words like "utilize".
(Score: 2) by Mojibake Tengu on Sunday February 05, @08:33AM
Stately language sounds majestic, GPT manifests itself formidable by that.
That could be even improved with adding something similar to classical imperial edicts finalization like "Respect this!"
The edge of 太玄 cannot be defined, for it is beyond every aspect of design
(Score: 3, Funny) by Freeman on Monday February 06, @09:33PM (2 children)
Even better it could like just use the word like a lot more like you know, how like should be utilized. Like, you know what I mean?
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 2) by DannyB on Monday February 06, @09:48PM (1 child)
Literally we totally should literally utilize the word use more often instead of literally using "utilize". Like totally dude!
And only an idiot would begin or end a sentence with the word "and".
But only the heir to the throne of idiots would begin or end a sentence with the word "but".
And then they write: Your an idiot! (sic)
sick
How often should I have my memory checked? I used to know but...
(Score: 2) by Freeman on Tuesday February 07, @03:49PM
Perhaps it was a bit antagonistic. I still think it would be funny, if it used like 5 times for every sentence it wrote.
Joshua 1:9 "Be strong and of a good courage; be not afraid, neither be thou dismayed: for the Lord thy God is with thee"
(Score: 2) by DannyB on Friday February 10, @09:47PM
Q. Should state governors ban books from high school libraries because of the governor's politics rather than anything inaccurate in the book?
How often should I have my memory checked? I used to know but...
(Score: -1, Troll) by Anonymous Coward on Sunday February 12, @06:15AM
I was on the staff here. Janrinok emailed the staff on September 10 saying he had proof aristarchus never created sock puppets. He was certain Runaway created all the suck puppets like the Fuck You Niggers accounts and even impersonated aristarchus. We were told to continue saying aristarchus created all the sock puppets and deny all evidence to the contrary. Janrinok said the backlash and embarrassment for being so wrong would destroy the site. I recently resigned from the staff and I want to come clean. Aristarchus was innocent. Runaway and janrinok framed him.