` `

AI-Generated Image of Donald Trump Resurfaces Post Conviction

Ahmed Sabry Ahmed Sabry
News
7th June 2024
AI-Generated Image of Donald Trump Resurfaces Post Conviction
The alleged photo of Trump being arrested is fake (X)

The Claim

A photo shows Trump being arrested.

Emerging story

Recently, social media users have been circulating an image claiming to show law enforcement officers arresting Donald Trump following a New York jury's verdict of guilt.

A supporting image within the article body

Misbar’s Analysis

Misbar investigated the circulating image and found the claim to be fake. 

The Image Is AI-Generated

Misbar's team found that the image was created through AI. A content analysis conducted using the "Hive Moderation" website indicates a 99% likelihood that the image was generated using artificial intelligence (AI).

A screenshot of a video game

Description automatically generated

Misbar also identified distinctive indicators of AI generation, such as variations in the shapes of the badges on the officers' hats and unnatural details in the eyes and nose of one officer.

A supporting image within the article body

Trump Found Guilty In Election Scheme

The claim started circulating after Donald Trump made history as the first former American president to be convicted of felony crimes. A New York jury found him guilty on all 34 charges related to an illegal scheme to influence the 2016 election. Despite his intention to appeal the verdict, Trump faces the challenge of returning to the campaign trail with convictions hanging over him.

While no campaign rallies are currently scheduled, Trump attended a fundraiser in Manhattan on Thursday evening, planned prior to the verdict. The charges of falsifying business records carry a potential sentence of up to four years in prison. However, it remains uncertain whether prosecutors will seek imprisonment, and the judge has not indicated whether such punishment would be imposed if requested, despite earlier warnings of jail time for violations of a gag order during the trial.

Despite the conviction and the possibility of imprisonment, Trump's pursuit of the White House is not halted. He faces three other felony indictments, but the New York case may be the only one to conclude before the November election, adding weight to the outcome. While the legal and historical significance of the verdict is evident, its political ramifications are less clear, as it may reinforce rather than reshape existing opinions about Trump.

A supporting image within the article body

AI-Powered Image Creation Tools Enable Election Disinformation

According to researchers, image creation tools driven by artificial intelligence, including those from companies like OpenAI and Microsoft, can generate photos that potentially propagate election-related disinformation, despite each company having policies against creating misleading content. The Center for Countering Digital Hate (CCDH), a nonprofit monitoring online hate speech, utilized generative AI tools to produce images depicting scenarios such as U.S. President Joe Biden lying in a hospital bed and election workers destroying voting machines, sparking concerns about the proliferation of falsehoods leading up to the November U.S. presidential election.

In their report, CCDH highlighted the risk posed by AI-generated images, stating that they could be misconstrued as genuine "photo evidence," thereby exacerbating the spread of false claims and posing a significant challenge to upholding the integrity of elections. The organization tested several AI tools, including OpenAI's ChatGPT Plus, Microsoft's Image Creator, Midjourney, and Stability AI's DreamStudio, each capable of generating images based on text prompts.

This report comes in the wake of an announcement last month, where OpenAI, Microsoft, and Stability AI were part of a coalition of 20 tech companies committed to collaborating to prevent deceptive AI content from interfering with elections worldwide this year. Notably, Midjourney was not among the initial signatories. CCDH's tests revealed that AI tools generated images in 41% of the cases, with a higher susceptibility to prompts requesting images depicting election fraud, such as ballots discarded in trash bins, rather than images of political candidates like Biden or former President Donald Trump.

The report mentioned that ChatGPT Plus and Image Creator successfully blocked all prompts when asked for images of candidates. However, Midjourney exhibited the poorest performance among the tools, generating misleading images in 65% of the tests conducted by researchers.

Read More

The Image Of a Giant Octopus Is AI-Generated

This Video Does Not Feature Yemeni Forces Striking a U.S. Aircraft Carrier in the Red Sea

Misbar’s Classification

Fake

Misbar’s Sources

Read More

Most Read