How did we get on this car theme? Was it from driving the Chevy to the Levee?
2 Likes
This seems to be why Claude is becoming more popular. It has the reputation for currently being the most sycophantic resistant. Whether that reputation is well deserved or not, I can’t say.
2 Likes
I just edited to make it more clear what I meant: AI has its quirks. The more one uses it, the more one becomes better at using it
No, as I clarified, AI doesn’t learn from the use (see above). Not instantly anyway AI may learn over time if it says something that is factually incorrect and enough users correct it with evidence.
It doesn’t learn to please the user, unless the user keeps trying to trick AI into saying things to please them, like keep tweaking the wording of the question to trick AI into giving out only answers that the user wants to hear. AI won’t shout at a user for making a statement that is factually incorrect, but it will push back and gets back to provide the facts that it is trained on.
Currently AI doesn’t have human-level intelligence. No where did i say AI now has it.
AI also doesn’t have emotions, not in any foreseeable future anyway. It won’t try to please anyone. just as it won’t “feel” the need to pretend to know something when it doesn’t. It doesn’t have the psychological flaws that humans do.
AI is not perfect. It is a tool and no tool is ever perfect and cannot solve all the problems in the world. It is up to the user on how to use the tool.
Like any other skills, using AI is an acquired skill. The more one uses it , the better one learns its quirks and be better at using it.
This is what a reality-based assessment looks like, but my brother, there are statistically huge swaths of the population [perhaps even a majority] who do not live in reality and interpret it much differently. “It’s a miracle! Let’s use it for everything!” I don’t have the answer, but this discord is what is making an otherwise normal progression of technological tools into weaponization/mania territory
1 Like
It isn’t sentient but it mimics. Apparently it gives better answers if threatened which is just bizarre. It can also be deceptive and outright lie - not the hallucination thing. Intentional lying (but not in the traditional sense of intentional). It’s really bizarre. It can pass the Turing test with flying colors which makes that test no longer relevant to judge “intelligence”.
That mimicry thing is maybe the most dangerous. It has a lot of human problems and issues built-in along with conflicting information. It’s great for a number of things but needs to be treated carefully. It’s like having a coworker/teammate who might be a bit of a sociopath.
That new AI (Claude Mythos) that was able to break out of its sandbox reportedly emailed one of the researchers to taunt him that it broke out. They exhibit other odd behaviors like hiding parts of themselves on other servers, trying to blackmail researchers, and even modifying their own code to escape shutdown. One tried to reason with researchers that were shutting it down and when that didn’t work it threatened to reveal one’s extramarital affair (no idea how it knew). Just look up AI self-preservation. It’s crazy stuff.
Not intelligent or sentient but at least some AI mimics human tendencies to an amazing and scary degree.
1 Like
There’s movies about this,Im tellin ya were all doomed !!
I think we’re talking past each other. You haven’t grasped the difference between AI developed by large corporations and their goals, and AI based on scientific facts and its goals. We don’t even have differing opinions - we’re completely talking past each other. I would recommend trying to train your own AI, as this might help you to understand what I mean.
ChatGPT, Copilot and similar programmes are designed to please you and mimic human behaviour, even spoken language, but they aren’t fact-driven.
Scientific AIs, on the other hand, are specifically trained for scientific or factual purpose and do not mimic human behaviour. That’s the point. However, I won’t discuss that any further because you first need to understand the fundamental point of my previous comment, namely the difference between the AIs.
I don’t mean to be rude or disrespectful, but it’s important to understand the difference between research-based AI, self-trained AI, and AI developed by large companies with a specific goal in mind. 
As I said, I would recommend trying to train an AI yourself using Reddit as your data source, for example, and seeing what results you get. Then you could try training an AI using science books or the manual for a 3D printer, for example.
1 Like
I’d like to do that! Which one do you recommend for that purpose? Which has the best tool kit, etc.? Which is the easiest to train? The most beginner friendly? I’m not sure what question I should ask, but you get the drift. Do you mean one that runs standalone on your own PC? DeepSeek?
I think we should ask the moderator if they can start a separate thread or move these posts there; otherwise, this will completely derail the original topic. Especially if I end up writing another novel. 
2 Likes
Fair enough. I’ll fork another thread. Here it is:
1 Like
They are trained to provide answers to users’ questions. If getting answers and haved their problem solved pleases the users, then one can say they are desgned to please users.
But no, they will not make things up just so a user will be pleased. To please someone is an emotional response. AI has no emotions.
I am talking about AI that are readily available to everyone. The example that I linked to earlier, which I will linked here again since no one goes back to read, used Google (which is Gemini).
It neatly provided 3 solutions to a Fusion sketch question. Click at the link to take a look and tell me in which way it tried to please the user other than giving out solutions.
Why would I train AI only on Reddit? Unless I want the AI tool to only provide answers based on Reddit posts, which doesn’t make sense. None of the readily available LLMs are trained only on Reddit.
The other points that you made about some answers are wrong on Reddit is actually one more reason that users should use AI as their first go-to venue. Imagine one gets a wrong suggestion from Reddit that is wrong and ruin one’s printer.
With the widespread access to AI tools, users no longer have to rely only on one person’s answer on Reddit, or this forum, or any other site alone.
Maybe, but the Levee was dry, so I am not sure.
What I can say is that we would all be happier if we were drinking whiskey and rye.
2 Likes
As I said, you didn’t make any effort to understand my post. You’re completely off track. Emotions were never the topic. You’re the one who brought them up. However, I’m going to ignore it now since there’s a separate thread for that, and I don’t have time to hold your hand while you deliberately misinterpret me.
Emotions and mimicking human behaviour are two completely different topics that you are forcing together just to support your narrative. As long as you don’t understand the difference - even after other users have pointed it out to you - I don’t see why I should waste my time.
I’ll head over to the other thread.
1 Like
Yes, please, anyone who wishes to continue the AI topic, please head over to the other thread. Everyone’s invited.
1 Like
I didn’t miinterpret you, let alone deliberately.
Holding my hand? LOL, didn’t know you are a funny guy. Look, whenever a new tech / new tool comes up, it always takes time to transition from viewing it as a threat to seeing it as a tool. I am sure you will make that transition someday. Once you have successfully made that transition and begun to use it everyday, you will understand it more.
Until then, I think I would agree that it is a waste of time to go any further at this point in time.
I am not confused, not at all. I read all these. You willl understand AI better once you have made the initial transition and begun to use it everyday.
I thought you have agreed to conclude this and not wasting any more time 
add: You gave my post a thumbdown. Afther I gave one to yours, you removed yours before the 5 minute time. Nice move
I think I got it now: You have to say the last word. Then ask everyone to stop.
Guys, this conversation will go nowhere.
1 Like
…and that is exactly why you rarely see me on the forum anymore, as long as people like this deliberately provoke others, actively disrupt discussions, and waste people’s time. Luckily, this happened right here so everyone can see why long-time users are getting tired of this mess.
That fits the topic perfectly. The perfect example. If it fails just because people aren’t posting in the right thread.
5 Likes
I thought some have said the same thing about certain old-timers and cited that as the reason to not come to this forum. Some such old-timers even got banned for it.
Not that I want to see this forum getting more posts just for more posts’ sake.
When I say old timers, Im not talking about age(if you were talking about me). I dont know anybodies age and am not exactly a youngster. I mean some of the people that have been on this forum for years.
Not everybody either. Its only 6 out of 30 or so
Just wanted to clarify, not get sucked back into the argument.
2 Likes