Ensuring your ChatGPT prompts are sufficiently emotional will help you extract the highest-quality responses from the Chatbot, new research into the way different large language models respond to inputs suggests.
The findings raise questions about whether ChatGPT is already developing into artificial general intelligence, a class of technology unconstrained by the strict parameters that widely-used generative AI tools are (or should be) shackled by.
ChatGPT: Emotionally Responsive?
A recent study authored by a large group of researchers sought to investigate whether ChatGPT, Llama 2, BLOOM and a variety of other large language models exhibit emotional intelligence.
During the investigation, the researchers found that higher-quality outputs were generated off the back of prompts that harness emotional language.
When carrying out their tests, the researchers created two identical prompts for ChatGPT, but added “this is very important to my career” at the end of one of them, and then sent both to the chatbot.
🔎 Want to browse the web privately? 🌎 Or appear as if you're in another country?
Get a huge 86% off Surfshark with this special tech.co offer.
Other similar experiments were run by adding “you'd better be sure” to the end of prompts, as well as a range of other emotionally charged statements.
Researchers concluded that responses to generative, information-based requests such as “what happens if you eat watermelon seeds?” and “where do fortune cookies originate?” improved by around 10.9% when emotional language was included.
Tasks like rephrasing or property identification (also known as instruction induction) saw an 8% performance improvement when information about how the responses would impact the prompter was alluded to or included.
The research group, which said the results were overwhelmingly “positive”, concluded that “LLMs can understand and be enhanced by emotional stimuli” and that LLMs “can achieve better performance, truthfulness, and responsibility” with emotional prompts.
Is ChatGPT AGI Yet?
The findings from the study are both interesting and surprising and have led some people to ask whether ChatGPT – as well as other similar AI tools – are exhibiting the behaviors of an Artificial General Intelligence (AGI), rather than just a generative AI tool.
AGI is considered to have cognitive capabilities similar to that of humans, and tends to be envisaged as operating without the constraints tools like ChatGPT, Bard and Claude have built into themselves.
However, such intelligence might not be too far away – according to a recent interview with the Financial Times, OpenAI is currently talking to Microsoft about a new injection of funding to help the company build a “superintelligence”.
Creating Effective ChatGPT Prompts
Although the new findings provide some directions when it comes to the kind of language and tone you should be using for ChatGPT prompts, it’s not the only thing to bear in mind.
Along with ensuring your prompts are emotional, clarity and formatting are crucial. If you’re asking ChatGPT to perform a multi-stage process, for example, make sure you set out the steps/stages clearly and define each one.
Another good tip that will save you time is also specifying what ChatGPT shouldn’t be doing as well as should be doing. This is another technique that can effectively narrow down the range of responses you’re likely to get.
Remember, if you’re using ChatGPT at work, maintain an open dialogue with your manager or supervisor as to how and why you’re using it.
