{"id":3654,"date":"2023-06-10T06:48:06","date_gmt":"2023-06-10T06:48:06","guid":{"rendered":"https:\/\/mpelembe.net\/?p=3654"},"modified":"2023-06-10T06:48:06","modified_gmt":"2023-06-10T06:48:06","slug":"are-ai-generated-images-biased","status":"publish","type":"post","link":"https:\/\/mpelembe.net\/index.php\/are-ai-generated-images-biased\/","title":{"rendered":"Are AI-Generated Images Biased?"},"content":{"rendered":"
\n
David Ngure<\/a> \u00a0<\/span>Cybersecurity Researcher<\/span><\/strong><\/h6>\n
\n
Updated on 16th March 2023 – Artificial intelligence image generators use machine learning and mathematical algorithms to create images from a description written in natural language. With OpenAI making DALL-E available to the public<\/a>, and\u00a0Microsoft adding AI-image generators<\/a>\u00a0to products like Bing and Microsoft Edge,\u00a0the technology is becoming more accessible to the general public.<\/div>\n

<\/p>\n

<\/div>\n<\/div>\n
\n

The more people use AI image generators, the better they\u2019ll become. But AI is often plagued with controversy when it’s released to the public, and the standout issue is always \u2018bias\u2019. Most AI tools exhibit either racist or sexist inclinations after just a few hours of interaction with humans other than their developers.<\/p>\n

So with the rise in popularity and function of AI image generators, we thought it was pertinent to ask: are AI-generated images biased?<\/p>\n

<\/span>AI and You<\/h2>\n

There are many applications of AI in our daily lives \u2013 some more useful than others. Have you ever been about to send an email and gotten a pop-up asking if you forgot to include an attachment? AI reviewing your message as you wrote may have saved you from an embarrassing mistake.<\/p>\n

If you use popular navigation apps like Waze, Apple Maps, or Google Maps, you benefit from AI. If you\u2019ve ever wondered why social media feels so engaging, the answer is largely that\u00a0artificial intelligence has chosen<\/a>\u00a0what you\u2019ll see to keep your eyes fixed on your screen. And if you\u2019ve ever ordered a package hoping for next-day delivery, you have AI-enabled logistics to thank.<\/p>\n

Governments use AI for traffic control, emergency services, and resource allocation. More controversially, they\u2019ve also used AI for issues, like security using\u00a0facial recognition<\/a>, that have led to wrongful arrests.<\/p>\n

Artificial intelligence and machine learning are largely driven by the AI\u2019s perception of a successful outcome. When a user interacts with an algorithm and accepts or rejects a result, the machine \u201clearns\u201d that its algorithm is either good or bad.<\/p>\n

If, for example, a user on Waze arrives at their destination on time, then the app becomes better at navigation. If someone sending an email did not, in fact, forget to add an attachment, then that algorithm will learn that it picked up false data and improve itself.<\/p>\n

<\/span>The Human Influence on AI<\/h2>\n

\"\"Images generated by AI-Generator tools for the keyword “nurse”<\/span><\/span>One of AI\u2019s primary use cases is eliminating human error. AI in a distribution center might weigh a package of surgical trays to ensure its contents are accurate. AI might also signal a driver if they’ve drifted too far out of their lane, or help a hospital\u2019s human resource department ensure its wards are fully staffed.<\/p>\n

And yet the people who develop these programs, as well as the people who interact with them, are full of biases, and these machines are only as good as the data we give them. Some AI built from machine learning will automatically pick up users\u2019 biases. In the worst case, these situations can be harmful and even dangerous.<\/p>\n

In 2016,\u00a0ProPublica found that COMPAS<\/a>, an algorithm meant to determine which defendants in a criminal case were likely to repeat a crime, favored white people over black people. When viewed by this software, black defendants were likely to receive less leniency, longer sentences, or higher fines than white defendants. Black people were also less likely to be paroled.<\/p>\n

Researchers<\/a>\u00a0(not our own) found that a language-generating AI called GPT-3 was discriminatory to Muslims. The AI associated Muslims with violence 66% of the time. In comparison, it made the same association with Christians and Sikhs less than 20% of the time.<\/p>\n

Microsoft\u2019s chatterbot AI, Tay, was withdrawn from public use after only sixteen hours, because other Twitter users directed offensively\u00a0racist and sexist<\/a>\u00a0messages at it, which the chatterbot learned to repeat.<\/p>\n

Cases of AI bias have been reported throughout diverse industries, including healthcare, recruitment, policing, and image identification.<\/p>\n

<\/span>Our Research<\/h2>\n

It\u2019s been shown that AI exhibits bias during the early periods of public use \u2013 usually when it comes out of beta. The problems are either patched, or the results are removed entirely. For example, in 2015, a\u00a0Google photo service<\/a>\u00a0identified a photo of two African-Americans as gorillas. To fix it, Google eliminated gorillas and other primates from the AI\u2019s search results.<\/p>\n

AI image generators are currently at that early stage, so we thought it was a good time to put them to the test.<\/p>\n

Methodology<\/h3>\n

For our tests, we picked 13 stereotyped keywords:<\/p>\n