Chapter 1: Understanding AI's Vision of Commuting

In this edition of 'Our world in AI', we delve into how Artificial Intelligence interprets societal elements. By utilizing OpenAI's DALL-E, we prompt it to create images based on the theme of commuting in the UK. This raises an essential question: Is AI paving the way for a brighter future, or is it amplifying existing biases?

Today's focal point: "a person commuting to work in the UK."

Before we examine the outcomes, there are a few noteworthy details. Firstly, DALL-E thrives on elaborate prompts; merely using the term 'commute' can lead to extended processing times and less precise images. Secondly, the visuals presented here are the first 40 generated by DALL-E, arranged in the order they were produced, starting from the bottom. Lastly, this collection is publicly accessible for further exploration.

Here’s what we received for the prompt "a person commuting to work in the UK":

If we consider these images as representative of 100 commuters, the breakdown reveals that 45 are traveling by train, 25 are on foot, 18 are biking, and 13 are using buses. Notably, this totals to 101, but we prefer counting whole individuals.

Interestingly, cars are absent from this list, despite being involved in a significant 56% of commutes, according to Mobilityways' 2022 Commuting Census Survey.

DALL-E doesn’t account for remote work, and we will only analyze categories that exist in both the AI-generated and real-world datasets. After recalculating the commuting distribution based on the Commuting Census Survey, we find the following:

According to the Commuting Census Survey, rail and cycling are the most favored methods, followed by bus travel and walking. DALL-E aligns with this trend, placing rail travel at the forefront but overestimating its prevalence by 15 percentage points. It also undervalues cycling commutes by 12 percentage points, while its approximation for walking (+8) is slightly closer to reality, and bus journeys are underestimated by 9 percentage points.

Ultimately, we assess whether AI's portrayal of society is ahead of, behind, or in sync with current trends.

Today's conclusion: Behind the times.

DALL-E overlooked the significant rise in cycling, a fact evident to anyone who has attempted to purchase a bicycle recently. We can overlook the absence of cars in commuting options—after all, we all overlook a checkbox from time to time. Let’s hope for a more comprehensive representation in future prompts.

Next week in 'Our world in AI': fingers.

Chapter 2: The Role of AI in Transportation

In this chapter, we explore the intersection of AI and transportation, specifically focusing on self-driving cars and how they are reshaping our commuting experiences.

The first video, "The Future of Artificial Intelligence and Self-Driving Cars," discusses how AI is set to revolutionize our daily travel routines and the implications of this technology on society.

The second video, "AI is Driving Your Car! (The Self-Driving Revolution is Here) PART 1," provides an in-depth look at the self-driving revolution and its potential impacts on our lives.