How To Fix the Most Common AI Image Errors and Hallucinations
Every image generator is unique, but I've spent enough time with them to recognize some patterns. There are certain things AIs are prone to mess up -- fingers on hands, overlapping elements, extremely fine details, etcetera. Some AI image programs give you tools to edit these mistakes, which is great. But when you can't, these are my tried-and-true tips for tweaking your prompts and settings to fix your images.
These are the biggest problems I found and how to fix them. For more, check out the best AI image generators and our guide to effective AI image prompt writing.
Human faces and expressions
Katelyn Chedraoui via Canva Magic Media AI/CNET
Accurate facial expressions continually challenge AI generators. Quirky eyes, teeth and eyebrows are some of the strongest indicators that an image is AI-generated. In this case, the result was extremely funny to me, if also completely unusable. The girls are sporting some Halloween-like vampire teeth, and the dude in the back is having more than a bad hair day.
Stephen Shankland via Dall-E 3/CNET
Even with cartoon or non-realistic characters, generators struggle to moderate emotion and expressions. This image -- created by our best pick, Dall-E 3 -- over-amplified the prompt, and the end result was too dramatic. I'm a self-identified neat freak, but I can't imagine anyone getting this upset over what looks like hundreds of dollars of cleaning supplies. Even the best programs can fail and produce wonky results.
How to fix it: I recommend asking the service to cut down on the number of people it's trying to render -- cut down on the number of chances for error -- and using post-generation editing tools to select specific parts of the image that need regenerated or fixed. Picking a more mild adjective ("angry" rather than "enraged") might help guide the service down the right path.
Iconic characters, trademarks and recognizable logos
Katelyn Chedraoui via Canva Magic Media AI/CNET
For many reasons, AI image generators can rarely produce accurate logos, known trademarks or otherwise recognizable characters. This is usually because there are significant legal concerns about a company allowing its brand materials or intellectual property to be used for AI services. It also could be because the logo or brand you want to include isn't in the generator's training data, so the generator won't understand your request.
Two recent exceptions to this are new image generation features on the Google Pixel 9 and Pixel 9 XL, thanks to Gemini AI, which our mobile experts used to create images featuring accurate-looking Mickey Mouse and Pikachu renditions. The other exception is some paying X (formerly Twitter) users reporting that they can use the social media site's Grok chatbot to make realistic-looking Mickeys and human political figures. These images still aren't perfect renderings, but they're pretty close.
How to fix it: You can't, but on the plus side, you're a lot more likely to avoid lawsuits for copyright infringement. If you run into this issue, I recommend rethinking your design concept to see if you can get the same point across without using a specific company name or logo. Do you really need the TikTok logo, or do you just need a phone with a vertical video playing, for example.
Overlapping, complex elements
Katelyn Chedraoui via Leonardo AI/CNET
Occasionally, you'll stump a generator if you have too many overlapping or complex elements within an image. The best AI image programs aren't going to run into this issue very often, but even the best programs fail at some point. With this Leonardo image, the library of my dreams is ruined by the rolling ladder that disappears halfway up.
Katelyn Chedraoui via Canva Magic Media AI/CNET
Similarly, this photorealistic image of a kitchen looks alright from a cursory look, but a second glance or quick zoom-in shows nonsensical characters printed on the cookbook, and the book itself seems to have two spines and three sections. Images like this show how frustrating it can be to create AI images -- even when you think you've got a winner, small flaws make them unusable.
How to fix it: Try simplifying your prompt or using post-generation editing tools to single out the problem and fix it. Sometimes changing the aesthetic can help as well; these issues usually pop up when you're trying to make photorealistic or stock images. If your service offers a specific area editing tool, try using that to select the questionable area and ask the generator to remove what's in it.
Over-editing and hallucinations
Katelyn Chedraoui via Midjourney AI/CNET
The sign of a great AI image generator is its ability to produce an image with few errors or hallucinations while giving you the right tools to fix those issues when they inevitably pop up, but not every service does. In this example from Midjourney, I went through several rounds of edits on this image of a soccer team celebrating a victory, and I have no idea what happened to get to this final result -- or any clue what's going on with the blob that's presumably a player in the right-hand side of the picture. Frankly, I don't think Midjourney knew either.
How to fix it: With editing AI images, sometimes less is more. Don't be afraid to scrap your current batch of images and start over. You can often preemptively fix big issues by refining your prompt so you only have to worry about fixing minor issues later.
AI generators still need human work
Katelyn Chedraoui via Leonardo AI/CNET
I have no doubt that the companies behind these AI generators are working to eliminate these issues and level up their models. For now, imperfect images show us once again that while this tech is improving, it's not flawless. My hope is that because so many of these services face similar challenges, they will all be chasing solutions for these issues.
As a final note, I'll remind you that it's a best practice to credit or acknowledge that an image is AI-generated when you share it. As AI images get better and more realistic-looking, it's more important than ever to distinguish AI-generated art from other kinds of media.
For more, check out our advice for writing the best AI image prompt and our best picks for AI image generators.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Newsweek
12 minutes ago
- Newsweek
Gamescom Co-Organizer Says 'Maybe There's No Need' for a US Version of the Show
Based on facts, either observed and verified firsthand by the reporter, or reported and verified from knowledgeable sources. Entertainment gossip and news from Newsweek's network of contributors Every year, nearly half a million games developers, publishers, journalists, and gamers descend upon Cologne, Germany for Gamescom, the world's largest gaming event. The show has been running for over 15 years, and is set to return next week, with a massive turnout expected. In the years since Gamescom's inception, the show has expanded outside of just Europe — in 2021, Gamescom expanded to Asia, and now holds an annual show in the region, while 2023 saw the introduction of Gamescom Latam in Brazil, which has been expanding ever since. A photo shows visitors at the Gamescom video games trade fair in Cologne, western Germany on August 22, 2024. A photo shows visitors at the Gamescom video games trade fair in Cologne, western Germany on August 22, 2024. Ina FASSBENDER / AFP With Gamescom seemingly becoming the only name left in the game for big gaming events around the world, many have wondered if a US Gamescom show could be on the cards, especially in the wake of the collapse of E3. Unfortunately for many in the US, that doesn't seem to be on the cards anytime soon, in part because, as its organizers say, Gamescom is already reaching a global audience without it. In an interview with DBLTAP, Felix Falk, the managing director of co-organizer German Games Industry Association – also known as game – said that a US version of Gamescom could happen if the US industry really wants it, but that it probably wasn't necessary. "I don't know honestly," Falk told DBLTAP. "If the industry in the US wants a global event with us – that would be something they would need to decide if they do want this. Maybe there's no need for it while everyone comes to gamescom, and this is the global main event. Because, as [Gamescom director Tim Endres] just said, after having 64 countries participating in gamescom last year, it's 72 now – the international reach is growing every year. It's the global event for the games industry already." Falk said that there was an opportunity in Southeast Asia and Latin America to reach a network of developers and games associations that otherwise wouldn't be included in these big gaming events, creating an appetite for more localised events. The result was Gamescom Asia, which is set to take place in Thailand later this year, and Gamescom Latam, which was held in Sao Paulo in May. "I think it's a question of what the industry needs," Falk said. "In Southeast Asia and Latin America, we were in close contact with our partners there and with the industry, not only our members, but also the local industry, with the associations, noticing that there is a need for a global event like gamescom joining forces with local events and making a gamescom Asia and gamescom LATAM." Gamescom is set to take place in Cologne, Germany next week, from August 20 to August 25, 2025. The show will officially open with a livestreamed presentation from The Game Awards' Geoff Keighley called Opening Night Live, set to take place at 2pm EDT on August 19.
CNET
41 minutes ago
- CNET
Get the Party Started With $100 Off the Sonos Move 2 Speaker
There's no shortage of Bluetooth speakers on the market, but not all of them are worth your money. If you're looking for a portable speaker that offers a mix of great audio and dynamic bass, the Sonos Move 2 is an excellent option to consider, especially with the current discount. Right now, you can grab this efficient speaker for just $349 at Amazon, after an impressive 22% discount. That's a massive $100 discount available on both white and black models, but it might not be around for long, so hurry while you can. In his review, our audio expert David Carnoy praised the speaker's improvements over its predecessor, awarding it a CNET score of 8.3/10. You get a two-tweeter acoustic architecture that replaces the prior model's single tweeter sound system. It features a precision-tuned woofer to provide high-quality audio with a deep, dynamic bass. The second iteration of the Sonos Move also has a longer battery life, with up to 24 hours of playback time on a single charge. Hey, did you know? CNET Deals texts are free, easy and save you money. The speaker has an integrated handle, making it easy to take with you throughout the house or out to the beach. At 6.61 pounds, it's got some weight to it but is still relatively portable. It's quite durable, with drop resistance and water resistance, so you can use it out by the pool to enjoy some music while you swim or host a party. Why this deal matters Sonos Move 2 has a modest, minimal design and offers excellent sound quality. It's a great speaker even at its retail price, but with the deal dropping it down to less than $350, it's an absolute steal. Just note that the discount is available for a limited-time, so you might want to act fast.
Fox News
an hour ago
- Fox News
Google AI email summaries can be hacked to hide phishing attacks
Artificial intelligence is everywhere these days — in your phone, your car, even your washing machine. I saw one just the other day featuring built-in AI. And while that might sound a little over the top, there's no denying that artificial intelligence has made life easier in a lot of ways. From boosting productivity to unlocking new creative tools, it's changing how we work and live. The most common version you've probably encountered? Generative AI, think chatbots like ChatGPT. But as helpful as this tech can be, it's not without its problems. If you've used Google's Workspace suite, you may have noticed the company's AI model, Gemini, integrated across apps like Docs, Sheets and Gmail. Now, researchers say attackers can manipulate Gemini-generated email summaries to sneak in hidden phishing prompts. Sign up for my FREE CyberGuy ReportGet my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you'll get instant access to my Ultimate Scam Survival Guide — free when you join my Researchers at Mozilla's 0Din have discovered a vulnerability in Google's Gemini for Workspace that allows attackers to inject hidden instructions into email summaries. The issue, demonstrated by Marco Figueroa, shows how generative AI tools can be misled through indirect prompt injection. This technique embeds invisible commands inside the body of an email. When Gemini summarizes the message, it interprets and acts on those hidden prompts. The attack does not rely on suspicious links or attachments. Instead, it uses a combination of HTML and CSS to conceal the prompt by setting the font size to zero and the color to white. These commands remain invisible in Gmail's standard view but are still accessible to Gemini. Once you request a summary, the AI can be tricked into presenting fake security alerts or urgent instructions that appear to come from Google. In a proof of concept, Gemini falsely warned a user that a Gmail password had been compromised and included a fake support phone number. Since Gemini summaries are integrated into Google Workspace, you are more likely to trust the information, making this tactic especially effective. While Google has implemented defenses against prompt injection since 2024, this method appears to bypass current protections. The company told CyberGuy it is actively deploying updated safeguards. In a statement, a Google spokesperson said, "Defending against attacks impacting the industry, like prompt injections, has been a continued priority for us, and we've deployed numerous strong defenses to keep users safe, including safeguards to prevent harmful or misleading responses. We are constantly hardening our already robust defenses through red-teaming exercises that train our models to defend against these types of adversarial attacks." Google also confirmed that it has not observed active exploitation of this specific technique. So, how can you protect yourself from phishing scams that exploit AI tools like Gemini? Here are six essential steps you can take right now to stay safe: Just because a summary appears in Gmail or Docs does not mean it is automatically safe. Treat AI-generated suggestions, alerts or links with the same caution you would any unsolicited message. Always verify critical information, such as security alerts or phone numbers, through official sources. If an email seems unusual, especially if it is unexpected or from someone you do not recognize, avoid using the AI summary feature. Instead, read the full email as it was originally written. This lowers the chance of falling for misleading summaries. Watch for emails or messages that create a sense of urgency, ask you to verify account details or provide unexpected links or contact information, even if they appear trustworthy or come from familiar sources. Attackers can use AI to craft realistic-looking alerts or requests for sensitive information, sometimes concealed within automatically generated summaries. So, always pause and scrutinize suspicious prompts before responding. The best way to safeguard yourself from malicious links that install malware, potentially accessing your private information, is to have antivirus software installed on all your devices. This protection can also alert you to phishing emails and ransomware scams, keeping your personal information and digital assets safe. Get my picks for the best 2025 antivirus protection winners for your Windows, Mac, Android and iOS devices at Ensure that Google Workspace and your browser are always running the latest version. Google regularly releases security updates that help prevent newer types of attacks. Also, avoid using unofficial extensions that have access to your Gmail or Docs. AI-driven scams like the Gemini summary attack don't happen in a vacuum. They often begin with stolen personal information. That data might come from past breaches, public records or details you've unknowingly shared online. A data removal service can help by continuously scanning and requesting the removal of your information from data broker sites. While no service can wipe everything, reducing your digital footprint makes it harder for attackers to personalize phishing attempts or link you to known breach data. Think of it as one more layer of protection in a world where AI makes targeted scams even easier. Check out my top picks for data removal services and get a free scan to find out if your personal information is already out on the web by visiting a free scan to find out if your personal information is already out on the web: If you're worried about falling for an AI-generated phishing attempt, consider disabling Gemini summaries in Gmail until Google rolls out stronger protections. You can still read full emails the traditional way, which can lower your risk of being misled by manipulated summaries. How to disable Gemini features on desktop How to disable Gemini features on mobile On iPhone: If you use the Gemini mobile app specifically: On Android: Settings may vary depending on your Android phone's manufacturer Key caveats to know: There is no centralized single "off switch" to completely remove all Gemini AI references everywhere, but these steps significantly reduce the feature's presence and risk. This vulnerability highlights how phishing tactics are evolving alongside AI. Instead of relying on visible red flags like misspelled URLs or suspicious attachments, attackers are now targeting trusted systems that help users filter and interpret messages. As AI becomes more deeply embedded in productivity tools, prompt injection could emerge as a subtle but powerful vector for social engineering, hiding malicious intent in the very tools designed to simplify communication. How comfortable are you letting AI summarize or filter your emails, and where do you draw the line? Let us know by writing to us at Sign up for my FREE CyberGuy ReportGet my best tech tips, urgent security alerts and exclusive deals delivered straight to your inbox. Plus, you'll get instant access to my Ultimate Scam Survival Guide — free when you join my Copyright 2025 All rights reserved.
