Nowadays blog authors and journalists use artificial intelligence quite a lot or they should. AI computers are programmed (‘trained’ in the language of AI) and for some it means that the computer’s most recent training is September 2021. That’s the first limitation.
Date limited
For one AI provider, Poe, you won’t receive any information from it that was discovered and written about over the past 2.5 years. All the most recent information available in studies or scientific developments are unknown to the AI platform I use. That’s a severe limitation. Here is what AI says about being ‘trained’:
I have been trained on a diverse range of internet text, with a training cutoff in September 2021. This means that my knowledge is based on information available up until that time. I don’t have access to real-time information or events that have occurred after my training cutoff. It’s always a good idea to verify information and seek up-to-date sources for the most current and accurate information, especially when it comes to rapidly changing fields or recent events.
AI chat box ‘Poe’
Opinions
Here is what the AI chat box I use says about AI having an opinion. I asked this question: “Why can’t you have an opinion about something?”
As an AI language model, I don’t possess personal opinions or beliefs. I don’t have subjective experiences or emotions like humans do. I can provide information, analysis, or perspectives on various topics based on the data I’ve been trained on, but these responses are generated objectively and without personal bias. My purpose is to assist and provide helpful and reliable information rather than express personal opinions.
Poe
Ethical and moral issues
I asked my AI ‘language model’ the following question: “Can you discuss ethical and moral issues?” Here is the response:
Yes, I can discuss ethical and moral issues. While I don’t have personal beliefs or opinions, I can provide information and analysis on various ethical and moral topics based on the data I’ve been trained on. Ethics and morality are complex fields of study that explore questions about what is right or wrong, good or bad, and how individuals and societies should behave. I can provide different ethical frameworks, perspectives, and considerations that are commonly discussed in these fields. However, it’s important to note that ethical and moral issues often involve subjective judgments and can vary depending on cultural, religious, or philosophical contexts.
Poe
My thoughts
AI’s discussion of ethical and moral issues will not be a reflection of their own views (opinions) as it has none. They won’t be original or unique thoughts but the thoughts of humans as reported and recorded on the internet and perhaps in books and newspapers etc.
AI can’t voice opinions on moral issues as the computer cannot have original thought. It is a very sophisticated and very quick regurgitator or information which it knits together in excellent summaries using faultless English. Great but limited. We need to be aware of these limitations.
