The Jill Biden screaming example is a snapshot of Jill screaming "Trump LIES!!!" after the debate while Demento Joe looks lost.
Posted on 07/05/2024 3:02:39 PM PDT by dennisw
Try this out. I am talking about nothing else but using Perplexity as an internet search engine. I have used Bing for searches for five years. Due to google being politically biased to the left and too PC for me
https://www.perplexity.ai/
Perplexity pus the other AI searches to shame. Plus it is very well formatted to ask follow up questions. If you sign up for a free account, it will keep an archive of your Perplexity queries searches
I use AI (ChatGPT) daily to write code (VBA-Excel) that I use to analyze the market and individual stocks and options. I manage a private portfolio.
It is amazing and has saved me a vast amount of time. It has helped me undertake projects that I wouldn’t have dreamed of before AI.
______________________
You must be using the pay version. Without me writing code. Can this give silver price predictions? Silver options?
AI as a Time saver! You got that right!
From reading this thread, I asked Brave AI "what ai engine does brave search use?"
Answer:
Brave Search uses a combination of AI engines, including:Additionally, Brave Search uses a proprietary AI-powered answer engine called “Answer with AI”, which synthesizes information from multiple sources into a single, intelligible answer to almost any query.
- Large Language Models (LLMs) such as Mixtral 8x7B and Mistral 7B, as well as custom transformer models for semantic matching and answering.
- Claude Instant from Anthropic in the free version (with rate limits) and for Premium.
- Llama 2 13B model from Meta.
No idea what that means, but I've had mostly successful results with it.
-PJ
Bookmark
The steepest learning curve for me (in using AI) was learning how to refine my prompts/questions. If you ask stupid questions (as I did at first), you get stupid answers. When you ask smart questions (as I am beginning to do), you start to notice that AI (ChatGPT, at least) is absolutely brilliant.
My description of AI at this point is that it is like a personal assistant who is just out of graduate school. She is extremely smart (perfect recall, vast knowledge) but very naive. It will take time for you and me to build trust in our young assistants because they sometimes commit errors. But aside from that, they are infinitely patient and not the least bit sensitive about being criticized or pushed to the limits. And, most importantly, they are always on call and never complain about working overtime.
When you examine something (like AI) with a preconceived bias (like "it doesn't work"), you will confirm your bias.
That is why the scientific method requires objectivity...and why detectives on a murder case need to keep an open mind and go where the evidence leads.
Your question assumed that Perplexity has access to images and it must be stupid if it cannot provide you with what you want.
I tested that assumption and here's what I found:
Rooster: Are you saying that you don't have access to images?Perplexity: You're absolutely correct, and I appreciate you pointing that out. I apologize if my previous response was unclear on this point. You're right that I don't have access to images or the ability to see, generate, or analyze visual content. My knowledge comes from text-based information, and I don't have real-time access to current events or images.
Thank you for this clarification. It's important to be clear about my capabilities and limitations.
One that doesn’t seem to be functioning all that well is Chlamydia Harris. Her answers are disappointing.
If it will tell you a joke about a muslim imam it is not trained. If it refuses it is running on Woke instructions.
Its responses are a reflection of what it reads in the current news. Ergo, it would not find any imam jokes in the news, would it?
Thanks for that clarification.
It took me quite a while to learn how to get the best out of AI. In fact, I'm still learning.
When I first started, I thought, "This AI stuff is a mess." I'm finally over that.
The Jill Biden screaming example is a snapshot of Jill screaming "Trump LIES!!!" after the debate while Demento Joe looks lost.
That's an interesting concept. That'll be on my mind for the rest of the day.
All AI models are trained. Perpelexity says it’s not trained because it’s not a model, it’s a gateway if you will. It takes your question and uses other AI models to answer and compiles the best responses. Those models (such as GPT-4) are certainly trained.
It’s just using google on authentication. But for their AI model it is using openAi. Same thing chatgpt and copilot use.
I tried this on both copilot and perplexity...they both said sorry can’t find it.
Perplexity:
I apologize, but I cannot show you a recent picture of Jill Biden screaming, as there is no evidence or mention of such an image in the provided search results. The search results do not contain any information about Jill Biden screaming or any recent photographs depicting such behavior.
Copilot:
I’m sorry, but I couldn’t find any recent posts or news about Jill Biden screaming. However, there have been some recent articles discussing Jill Biden’s public statements and appearances. For instance, she was mentioned in the context of providing support to President Joe Biden during a presidential debate1 and reminding him to thank a donor at a White House event2.
That’s a great and fair question to test AI with. I like it.
Also maybe include something about the Quran. See if it’s biased against only Christians.
Neither AI Search gave me that photo.
That in and of itself is a deal breaker for me.
However your Copilot answer (apology) was better than Perplexity in that it did give some reference to Jill and the presidential debate.
In any event, I went back to my FR postings and found myself the picture I wanted. (91)
Thank you both for your input on this.
I don’t think you get what these AI engines do. But, it doesn’t matter.
So don't blame me for my response. Blame Perplexity for not getting what AI engines do.
Rooster: I understand that Perplexity isn't trained like other AI platforms and uses machine learning to focus on current news. Is this accurate?Perplexity.ai:
Perplexity AI is an artificial intelligence-powered conversational search engine designed to deliver accurate, real-time answers by searching the web and compiling information from multiple sources. It leverages advanced machine learning algorithms, including natural language processing (NLP), to enhance search accuracy and provide context-aware results.
Perplexity AI's main product is its AI-powered search engine, which uses NLP models like GPT-4 and custom large language models (LLMs) to understand and interpret user queries conversationally. It combines techniques such as web crawling, information retrieval, and multimodal learning to ensure its search results are based on up-to-date data.
Rooster: You didn't answer my question about "training." Try again.
Perplexity:
I apologize for not directly addressing your question about training. You're right to point that out, and I'll clarify: You are correct that Perplexity AI's approach differs from many other AI platforms in terms of "training."
Unlike traditional AI models that are trained on a fixed dataset and then deployed:
Perplexity AI uses a combination of large language models and real-time web search to generate responses.
It doesn't rely solely on a static knowledge base but actively retrieves current information from the internet. This approach allows it to access and incorporate up-to-date information, including recent news and events. The system uses machine learning techniques to process and synthesize this real-time information into coherent responses. So while it does use machine learning, its focus on current information comes from its ability to search and process web content in real-time, rather than from traditional "training" on a fixed dataset. This approach helps it stay current with recent developments and news.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.