ΑΙhub.org
 

How to spot AI hype

by
19 November 2021



share this:
microphone in front of a crowd

Hype itself is not inherently bad, but it sets inflated expectations about the technology, drives unnecessary fears, and detracts from meaningful discussions about today’s technology. We’ve already talked about how to avoid creating hype, but it’s equally important to know how to recognise it.

So how can you spot it and avoid the hype train?

1. Sensational and absolute language

Examples: “AI will completely revolutionise healthcare / construction / agriculture ”, “X is the most important development in AI – ever”.

Words like “best”, “worst”, “ever” “never” can be useful as a shorthand, but lack nuance. If you see phrases like this (especially in a headline) remember that there is likely to be more to the story.

2. There are no limits!

Examples: “AI can now predict the weather with 100% accuracy.”

No one is perfect and neither is AI so any claims that don’t acknowledge its limitations are likely to be hype. Look out for words and phrases such as “may”, “might”, and “could” that indicate that there are limitations without directly specifying them.

3. Emotive language

Examples: “Terrifying future for millions of workers as AI takes over”

Descriptions that evoke strong emotions are attention grabbing and engaging, but are not an accurate reflection of the technology.

4. Anthropomorphism

Example: “Depressed security robot throws itself into fountain”

Describing AI in terms that would normally be used for humans or living things is inaccurate as it ascribes abilities (e.g. having feelings) that it is not capable of.

5. Leaning into biases

Example: “If you think robots will steal your job – you’re completely right”

People are more likely to pay attention to and accept information that aligns with their beliefs. To avoid being drawn in by hype, evaluate claims that align with your values as thoroughly as information that doesn’t.

6. Underlying motivation

Example: “‘AI is the future of housework’ says maker of robot vacuums.”

Think about why the source could be using hype: Are they trying to change your mind, or behavior, or sell you something? Who benefits from the hype?

7. Personalities over science

Examples: “AI Company X is a legacy of the genius of person Y”

While people are ultimately responsible for creating AI, discussions focused on people and personalities (both good and bad) distracts from the science itself.

8. Unrealistic or unrelated images

Examples: Sci-fi film robots, glowing blue brains

It’s true that representing AI visually can be difficult, but images of science fiction robots or fantastical images creates the impression of greater advancement than is realistic.

9. Vague claims

Examples: All of the above.

Headlines are brief by necessity, but the rest of the story should give more information about how the technology works. For example, explanations about the AI’s specific task, the data, algorithm, and hardware used.

10. If in doubt, check the original source

If you spot hype, or are unsure, one of the best things you can do is check an original source, the journal or conference proceedings. If you can’t find any, it is more likely that the story is all hype with no substance.

We hope you find this guide to spotting AI hype helpful, and you can also find all the tips in this PDF.



tags: ,


Joe Daly Engagement Manager for AIhub
Joe Daly Engagement Manager for AIhub




            AIhub is supported by:


Related posts :



What’s coming up at #AAAI2023?

Find out about the talks, workshops, tutorials, and other events scheduled at AAAI this year, taking place from 7 - 14 February.
03 February 2023, by

Science communication for AI researchers: our tutorial at #AAAI2023

Find out about our in-person hands-on course at AAAI.
02 February 2023, by

Riemannian score-based generative modelling

The winners of a NeurIPS 2022 best paper award write about their work on generative modelling.
01 February 2023, by

AIhub monthly digest: January 2023 – low-resource language projects, Earth’s nightlights and a Lanfrica milestone

Welcome to our monthly digest, where you can catch up with AI research, events and news from the month past.
31 January 2023, by

The Good Robot Podcast: featuring Abeba Birhane

In this episode, Eleanor and Kerry talk to Abeba Birhane about changing computing cultures.
30 January 2023, by

All questions answered: how CLAIRE shapes the future of AI in Europe

Watch the next in the series of CLAIRE's All Questions Answered (AQuA) events.
27 January 2023, by





©2021 - Association for the Understanding of Artificial Intelligence


 












©2021 - ROBOTS Association