Why does ChatGPT lack a personality? | Hacker News Reader

Why does ChatGPT lack a personality?

1 pointedisonsama20 comments

I really feel like ChatGPT needs a personality and it's so lame without one.

Does anyone have an idea why this doesn't exist?

It is instruction-tuned with very flatly-written responses. That makes the structure of the response more informative and less playful.

Personally I like the no-bullshit approach, but you could also fine-tune a personality if you wanted.

Sure I think that makes sense within reason. I also think that many would want one that has more of personality and I am wondering why no one has built this

They have. Try some of the LLaMA roleplaying finetunes, like Chronos:

https://huggingface.co/models?sort=modified&search=chronos

Or Samantha

https://huggingface.co/models?sort=modified&search=saman

Don't you think this would be more useful than ChatGPT? For example look at SmarterChild from years ago. It felt more genuine

If you are saying OpenAI needs more "personality" finetunes, you are absolutely right.

What do you mean when you say "personality" finetunes. I am not referring to a replika like feature. I mean something more like a mixture of heypi.com + smarterchild + chatGPT.

> Don't you think this would be more useful than ChatGPT?

In my experience, it just makes it harder to extract useful information.

I see what you are saying and that "can" be true. However everyone has diverse learning styles.

Some are more visual and others have learning disabilities. I think that ChatGPT is too uniform and if it was fine-tuned towards you personally that would be the best version

I believe it's the way OpenAI has positioned its model.

By itself, the model follows a straightforward and concise 'answer the question' approach. However, if you want it to be creative, scientific, or empathetic, you can engineer your prompts accordingly (bear in mind, it may spew out literal garbage while trying to answer certain advanced questions scientifically).

It can also enact prominent figures so that the responses are tailored according to how the actual person would answer.

This is the thing. I don't want to deal with prompt engineering. I want ChatGPT to be fine-tuned to me. My learning style, my unique experiences and what ultimately makes me a human.

That enables me to have a more meaningful way to collaborate. I don't want a version of google that summarizes information into a chat based interface with some "faux-human" responses.

It doesn't lack a personality. It just so happens that the personality is so close to your ordinary daily experiences that you don't perceive anything off. Try asking someone from a culture very different from yours.

This is not an answer. What do you define as personality? I don't have any experiences in my daily life that mirror even the positives of ChatGPT.

I.e a person who has a wide depth and breadth of knowledge that can be called upon at a whim.

When I refer to personality I mean that ChatGPT makes my life worse since it doesn't understand me or my experiences. I want real-time fine-tuning where ChatGPT actually evolves to fit me personally and I want this for every user

OK, I think I see the problem with this conversation. When people talk about personality in communication, they generally mean the manner of expression. What you seem to want is a tailoring of the content to your needs. At the moment, I believe the only way to do this is with prompt engineering. I suppose eventually some company will offer the ability to customize your account profile so that responses are better tailored.

Yeah I think you are right in this regard. Regarding the miscommunication. However I think it should be a mixture of tailoring content to your needs and having some positive expression that feels more human and intimate without a romantic component. Like a trusted colleague.

For the output, you can request for the results to be in the style of your favorite writer. It doesn't always work but try different styles to see what works.

> It doesn't always work.

This is the bummer. I want consistency and I don't care about it being in the voice of my favorite author. I personally don't have a favorite. I want it to be fine-tuned on me personally so the collaboration is more meaningful

You can add a personality through prompt engineering, even moreso with system prompt engineering via the API.

This is not what I want. I want prompt engineering to go away. Just fine-tune on me personally

What do you mean, no personality? ChatGPT sounds exactly like people I know working in PR.

Ok so a mixture of useful and useless information merged into an overwhelming output?

Reply on news.ycombinator.com