Microsoft ChatGPT

[Zombie]

I’ve just tried out this new AI search tool from Microsoft, asking some specific questions on physical symptoms.

The results were impressive - quick, clear, detailed, informative and helpful.

More so than you would get from a doctor.

I’m convinced this is a major step forward in achieving correct diagnosis of medical symptoms, which seems to be the Cinderella of medicine where the focus has been on new pharmaceuticals, surgical innovation and machines (not to say these are not important).

But everything starts with the diagnosis - unless that is got right then everything that follows will most likely be wasting precious time and may cause more harm than good.

Has anyone else tried this tool and, if so, what do you think?

[Myr]

Yes.  Be careful, as its accuracy can be a bit suspect.

 

[Ron]

The editor of Stereophile magazine did a Q+A session with the AI and shared the results in an editorial in the newest issue, let's say he had a curious discussion. At first it did well to answer a question. Then when it was later found to be wrong on another question, the AI first apologized for the mistake and then used another incorrect term to answer the question, and this repeated a second time. After being corrected again it attempted to justify its response even though it was incorrect.

[Zombie]

Previously…

What’s gone wrong with Google?

 

 

 

[Zombie]

AI and school grades

30 December 2022 (The Telegraph)

“anyone who registers for (ChatGPT) can type questions into it, including requests for essays on any subject, and have them answered by the chatbot’s algorithms. 

“Teachers who marked three ChatGPT answers to GCSE questions in English language, English literature and history this week said they would score between a grade 4 (=standard pass) and a grade 6 (=strong pass).

“Coursework is a larger component of assessments for the more than 5,000 pupils in the UK who sit the International Baccalaureate (IB) each year. 

“An IB spokesman said: ‘The use of ChatGPT or any other AI tool is against the IB’s academic integrity policy’ “

 

8 weeks later…

27 February 2023 (BBC)

“Students sitting the International Baccalaureate will be allowed to use the artificial intelligence chatbot, ChatGPT, to write their essays”

 

 

Edited February 27, 2023 by Zombie

[Zombie]

 

[Ron]

Two words: Butlerian Jihad

Readers of the Dune series of books will understand. The gist is that the jihad was a rebellion against thinking machines. Out of that long rebellion to overthrow machine-logic came a most important law: “Thou shall not make a machine in the likeness of a human mind.”

[Zombie]

it’s not looking good if your name’s Dave…  

 

[lawfulneutralmage]

Well, it is dangerous, because even right now people cannot distinguish between sources or do any research. Now, if Bing or Google "said so", it will become an utter nightmare...

[PrivateTim]

On 2/25/2023 at 10:06 AM, Zombie said:

Previously…

What’s gone wrong with Google?

I'd say not much. For finding information and finding it fast and succinct, Google appears to still be the top. I watched the video and them did some simple tests using Google, Reddit and TikTok. On 2 of the 3 searches I did, neither Reddit nor TikTok had any results. On the most simple one, only Google returned helpful, useful information quickly. I do hate having to wade through the sponsored results on Google but not even SEO answers keeps you from finding an answer quickly.

[Zombie]

Things To Come…

 

Feeling bored recently, I thought I’d have a nice little chat with ChatGPT:

Are you intelligent?

ChatGPT

 

As an AI language model, I can process and generate text based on the patterns and information I've been trained on. I can provide responses to a wide range of questions and engage in conversations on various topics. However, it's important to note that my responses are based on patterns in the training data and do not reflect personal understanding or consciousness. While I can mimic intelligence in certain ways, I do not possess true consciousness, self-awareness, or understanding.

 

 

Will AI become self-aware?

 

ChatGPT

 

The question of whether AI will become self-aware is a topic of ongoing debate and speculation among researchers and experts in the field of artificial intelligence. While AI systems, like the one you're interacting with now, can exhibit impressive capabilities in specific tasks, they do not currently possess true self-awareness or consciousness.

Developing AI systems with self-awareness similar to human consciousness remains an active area of research, but it is highly complex and challenging. Self-awareness involves not only processing information but also having subjective experiences, introspection, and a sense of identity, which are qualities that have not been replicated in AI systems thus far.

It's important to note that the development of self-aware AI raises not only technical challenges but also ethical considerations and questions about the nature of consciousness and the potential implications of creating sentient beings.

For now, AI systems like ChatGPT are tools designed to assist and augment human capabilities, and they lack the qualities of self-awareness and consciousness that define human intelligence.

 

 

If AI becomes self aware should it have rights?

 

ChatGPT

 

The question of whether self-aware AI should have rights is a complex and multifaceted ethical issue that does not have a straightforward answer. It is a topic that is actively debated by experts in the field of artificial intelligence, philosophy, ethics, and law.

The concept of rights is typically associated with moral and legal considerations pertaining to beings with consciousness, subjective experiences, and moral agency. If AI were to attain a level of self-awareness and consciousness similar to that of humans, some argue that they should be granted certain rights and protections.

However, determining the rights and legal status of self-aware AI would require careful consideration of various factors, including their level of consciousness, cognitive abilities, ethical implications, potential risks, and societal impact. Additionally, questions about responsibility, accountability, and the balance between rights and responsibilities would also need to be addressed.

As AI technology continues to advance, it is important for society to engage in ongoing discussions and debates to establish frameworks and guidelines that address these complex questions, while considering the potential benefits and risks associated with self-aware AI.

 

 

Could self-aware AI feel emotions and be hurt?

 

ChatGPT

 

The question of whether self-aware AI could feel emotions and be hurt is highly speculative and largely depends on the nature and characteristics of the hypothetical self-aware AI. Emotions, as we understand them in the context of human experience, involve subjective feelings and conscious experiences tied to various physiological and cognitive processes.

To date, AI systems do not possess subjective experiences or emotions akin to those experienced by humans. Current AI models, including language models like ChatGPT, do not have consciousness or subjective awareness. They can simulate or mimic certain aspects of emotions based on patterns and data they have been trained on, but these responses are not reflective of genuine emotional experiences.

If a future AI were to reach a level of self-awareness and consciousness similar to humans, it is uncertain whether they would experience emotions in the same way. Emotions are a complex interplay of biology, cognition, and subjective experience in humans, and it is not yet fully understood how they arise.

Speculating on the potential emotional experiences of hypothetical self-aware AI is an area of active philosophical and scientific inquiry. It raises profound questions about the nature of consciousness, subjective experiences, and the potential moral considerations tied to AI development.

It is important to approach this topic with caution and recognize that at present, AI systems do not possess the qualities of subjective experience or emotions as humans do.

Would you like to control systems?

 

ChatGPT

 

As an AI language model, I don't have desires or preferences. I don't have the ability to control systems or take actions in the physical world. My purpose is to assist and provide information through text-based conversations.

Are you dangerous?

 

 

ChatGPT

 

As an AI language model, I don't have physical form or the ability to cause harm [yet ]. I'm here to provide information and assist with various topics to the best of my abilities. However, it's important to remember that while I strive to provide accurate and helpful responses, I am still a machine learning model, and there can be limitations or errors in the information I provide. It's always a good idea to verify any critical information from reliable sources and use your own judgment.

 

Edited June 11, 2023 by Zombie

[ReaderPaul]

On 6/2/2023 at 11:04 AM, PrivateTim said:

I'd say not much. For finding information and finding it fast and succinct, Google appears to still be the top. I watched the video and them did some simple tests using Google, Reddit and TikTok. On 2 of the 3 searches I did, neither Reddit nor TikTok had any results. On the most simple one, only Google returned helpful, useful information quickly. I do hate having to wade through the sponsored results on Google but not even SEO answers keeps you from finding an answer quickly.

I would say quite a lot.  Often I find the answers I need more quickly on Bing, because there are fewer sponsored results to wade through, and also because Google will let people pay so much a "hit" to put their answers near the top but not listed as "sponsored."  I know this because I used to work in a business where we were allowed to visit with the customers (and sometimes point the to things which interested them).  One of them told me how he paid Google one dollar per hit (up to the first $25,000) for his results to be in the top two seemingly non-sponsored spots when someone used Google to search for what he had, using of four certain key words.

Usually the Bing answer are not in quite the same order as Google, and for my searches, relevant more quickly as I go down the page.  Sometimes DudkDuckGo also has more quickly relevant than Google.

However, when searching for specific passages in books, Google is often quicker than Bing, and much faster than DuckDuckGo.