The Conversation

Are you “AI fluent”? Whether we like it or not, being able to proficiently work with generative AI systems is increasingly becoming a job requirement at many companies – despite the fact that data on whether artificial intelligence actually boosts productivity is not at all definitive.

If you’ve ever been let down by a generic, anodyne or downright false answer from a chatbot, you’re far from alone. But Sandra Peter and Kai Riemer from the University of Sydney argue that it’s because you haven’t given the AI enough to work with.

“Without specific guidance, it’s like walking into a restaurant and asking for something good. You’ll likely get the chicken,” they write.

So how do we improve our prompts to get what we want? There’s a cute acronym for that: CATS. Include the context, angle, task and style, and you’re bound to get these prediction machines to spit out something useful.

Signe Dean

Science + Technology Editor
The Conversation Australia

Lead Story

What makes a good AI prompt? Here are 4 expert tips

Sandra Peter, University of Sydney; Kai Riemer, University of Sydney

Being ‘AI fluent’ is quickly becoming as important as being proficient in office software once was.

Ethics

The forgotten 80-year-old machine that shaped the internet – and could help us survive AI

Martin Rudorfer, Aston University

Leading American engineer and scientist Vannevar Bush thought that the memex was the answer to information nightmare of the 1940s.

Education

University students feel ‘anxious, confused and distrustful’ about AI in the classroom and among their peers

Elise Silva, University of Pittsburgh

Whether students and faculty are actively using AI or not, it is having significant interpersonal, emotional effects on learning and trust in the classroom.

Science

AI will soon be able to audit all published research – what will that mean for public trust in science?

Alexander Kaurov, Te Herenga Waka — Victoria University of Wellington; Naomi Oreskes, Harvard University

An AI audit of scientific research would likely expose some fraud and widespread inconsequential work. But we need to be careful it doesn’t discredit science in general.

AI and Humanity

Can AI think – and should it? What it means to think, from Plato to ChatGPT

Ryan Leack, USC Dornsife College of Letters, Arts and Sciences

Whether AI can ‘think’ is a different question than whether it is ‘intelligent.’

Quote of the week 💬

  • "The common narrative around AI and productivity is that AI automates mundane tasks, making us faster at doing things and giving us more time for creative pursuits. This, however, is a naive view of how work happens."

    – Jon Whittle of CSIRO in his article Does AI actually boost productivity? The evidence is murky

More from The Conversation