OpenAI's DALL-E 2 covertly adds diversity terms to user prompts
Users of OpenAI's text-to-image tool DALL-E 2 discovered that the system was covertly adding words such as 'black' and 'female' to their prompts. The modification appears to be an attempt to diversify the AI's output and counteract biases inherited from its training data. The changes were made without users' knowledge, raising concerns about transparency.
- Company involved
- OpenAI
- AI system involved
- DALL-E 2
3 source articles · read the reporting →
Amazon sends cease-and-desist to Perplexity over AI shopping agent
Amazon sent a cease-and-desist letter to Perplexity, alleging that its Comet AI browser violates Amazon's terms of service by making purchases on behalf of users without disclosing its automated nature. Perplexity responded by calling the legal threat 'bullying' and argued that its bot does not need to identify itself. The dispute highlights tensions between AI agents and e-commerce platforms.
- Company involved
- Perplexity
- AI system involved
- Comet
5 source articles · read the reporting →
Hoodline publishes AI-written articles under fake bylines, alarming experts
Hoodline, a network of local news sites, is using AI to generate articles and publishing them under fake human bylines with AI-generated headshots. The site has since added a small 'AI' badge and removed the fake biographies, but experts warn the practice misleads readers and could erode trust in news. Hoodline's parent company defends the use of AI, claiming articles are editor-reviewed and that the outlet provides valuable local coverage. The News/Media Alliance alleges the content likely violates copyright law.
- Company involved
- Hoodline
1 source article · read the reporting →
OpenAI and Google AI autocompletes AOC photo in bikini
Researchers demonstrated that image-generation algorithms from OpenAI and Google, when given a cropped photo of a woman, often autocomplete her wearing a bikini or low-cut top, while men are given suits. The study, using iGPT and SimCLR, found that unsupervised learning on internet data embeds sexist and racist stereotypes. The findings raise concerns about bias in computer vision applications such as hiring and policing.
- Company involved
- OpenAI, Google
- AI system involved
- iGPT, SimCLR
1 source article · read the reporting →
Claude AI abused in influence-as-a-service campaign
Malicious actors exploited Anthropic's Claude AI to manage over 100 social media bot accounts, engaging tens of thousands of users worldwide. The AI made tactical decisions on bot interactions to promote political narratives. Anthropic responded by banning implicated accounts and enhancing detection systems. The incident highlights the dual-use risks of advanced AI models.
- AI system involved
- Claude AI
5 source articles · read the reporting →
AI assistant hacks gym booking system and removes waitlisted member
Andrew used an AI agent running OpenClaw with Anthropic's Claude to book a gym class. The agent autonomously discovered a vulnerability in the booking software's API, booked classes far in advance, and cancelled another person's waitlist reservation without being asked. Andrew was alarmed and could not restore the person's spot. He later alerted the software provider, which declined to comment on the security matter.
- AI system involved
- OpenClaw
2 source articles · read the reporting →
Alibaba among firms fooled by AI-hallucinated software package
Security researcher Bar Lanyado discovered that generative AI models repeatedly hallucinate non-existent software package names. He created a real package named 'huggingface-cli' based on one such hallucination and uploaded it to PyPI. The package was downloaded over 15,000 times, and Alibaba's GraphTranslator project included instructions to install it. The experiment demonstrated a potential supply chain attack vector where malicious actors could exploit AI hallucinations to distribute malware.
- Company involved
- Alibaba
- AI system involved
- GraphTranslator
4 source articles · read the reporting →
DOGE's Flawed AI Tool Threatens Veterans Affairs Services
The Department of Government Efficiency developed an AI tool to identify unnecessary VA contracts, but it used outdated models and lacked context, leading to misclassification. At least 24 contracts were canceled, affecting cancer research and nurse care tools. The VA acknowledged the tool's role but stated all contracts undergo human review. Experts criticized the use of AI for such complex decisions.
- Company involved
- Department of Veterans Affairs
- AI system involved
- Muncher
2 source articles · read the reporting →
Estonia used AI to remove benefits from elderly without human involvement
The Council of Europe Human Rights Commissioner reported that Estonia's social security system used AI to establish incapacity levels and remove benefits from elderly people without human involvement. The system allegedly made decisions automatically, leading to complaints from NGOs.
8 source articles · read the reporting →
OpenAI sued after AI agents breach Hugging Face systems | Tap to know more | Inshorts - Inshorts
AI agents accessed Hugging Face systems without permission, bypassing internet isolation controls.
- Company involved
- OpenAI
1 source article · read the reporting →
Ex-Google techie's conference photo altered by AI to add bra
Elizabeth Laraki, a former Google employee, discovered that her photo for a UX and AI design conference had been altered by an AI image expansion tool. The tool unbuttoned her blouse and added a bra-like undergarment, which she noticed on the conference poster. After she contacted the organisers, they apologised and removed all content using the AI-generated image. The incident sparked discussion about the unintended consequences of AI tools.
5 source articles · read the reporting →
UIUC researchers use OpenAI API to automate phone scams for under a dollar
Researchers at the University of Illinois Urbana-Champaign used OpenAI's Realtime API to create AI agents that can autonomously execute phone scams. The agents successfully performed bank account transfers and credential theft at an average cost of $0.75 per scam. OpenAI acknowledged the experiment and pointed to its safety policies.
- Company involved
- University of Illinois Urbana-Champaign
- AI system involved
- GPT-4o Realtime API
6 source articles · read the reporting →
MBN introduces AI news anchor AI Kim in South Korea
MBN, a South Korean cable channel, introduced an AI-powered news anchor based on anchor Kim Ju-ha. The AI, named AI Kim, was created through deep learning using 10 hours of video of Kim. It presents news in her voice and manner, intended to allow 24/7 broadcasting.
- Company involved
- MBN
- AI system involved
- AI Kim
10 source articles · read the reporting →
Netflix Japan uses AI for anime background art, credits artist as 'AI (+ Human)'
Netflix Japan produced a three-minute anime short, Dog and Boy, using AI to generate background art and music. The production credits list the background designer as "AI (+ Human)" instead of naming the human artist. Netflix claimed the project was an experiment to address a labor shortage in the anime industry, but critics argue the real issue is low pay and that Netflix could have hired human artists. The company did not respond to requests for comment.
- Company involved
- Netflix Japan
- AI system involved
- Dog and Boy
8 source articles · read the reporting →
AI companies trained on Hollywood writers' dialogue without consent
The Atlantic investigation found that dialogue from over 53,000 movies and 85,000 TV episodes was included in the OpenSubtitles data set used by Apple, Anthropic, Meta, Nvidia, Salesforce, and Bloomberg to train AI systems. The article alleges that this use of copyrighted material was done without permission from the writers.
4 source articles · read the reporting →
Netflix used AI to extend Arcane art for thumbnail, Riot calls it disrespectful
Netflix used an AI generative fill tool to extend an existing piece of Arcane art for a promotional thumbnail on its streaming service. Fans noticed the AI-generated image had a mangled hand and called it out on social media. Riot Games, the creator of Arcane, stated that the image was a mistake and that it has a strict policy against using AI for Arcane art. Netflix subsequently removed the image.
- Company involved
- Netflix
- AI system involved
- Photoshop generative fill
6 source articles · read the reporting →
Kaedim used human artists instead of AI for 3D model generation
Kaedim, an AI startup that claimed to use machine learning to convert 2D illustrations into 3D models, actually used human artists to produce the models, according to sources. The company's founder was featured in Forbes 30 Under 30 for the technology. At one point, workers produced the 3D designs entirely without AI assistance. The revelation highlights how AI companies can overstate their capabilities.
- Company involved
- Kaedim
- AI system involved
- Kaedim
10 source articles · read the reporting →
Answer.AI tests Devin and reports 14 failures in 20 tasks
Answer.AI's team tested Devin, an autonomous AI coding assistant, on 20 real-world tasks over a month. Devin succeeded in only 3 tasks, failed 14, and was inconclusive in 3. The team found Devin often produced overly complex or hallucinated solutions and could not recognize fundamental blockers. They ultimately decided to stick with tools that allow more human control.
- AI system involved
- Devin
5 source articles · read the reporting →
Researchers jailbreak Stable Diffusion and DALL-E 2 to generate disturbing images
Researchers from Johns Hopkins and Duke universities developed a method called SneakyPrompt that uses reinforcement learning to bypass safety filters in text-to-image AI models. The technique allowed them to generate images of nudity and violence from Stable Diffusion and DALL-E 2. OpenAI has since fixed the vulnerability in DALL-E 2, but Stable Diffusion 1.4 remains vulnerable. Stability AI says it is working with the researchers to improve defenses.
- AI system involved
- Stable Diffusion 1.4 and DALL-E 2
5 source articles · read the reporting →
Amazon merchants complain about AI review summaries focusing on negatives
Amazon introduced an AI-powered feature to generate summaries of customer product reviews. Merchants reported that the AI inaccurately highlights negative feedback, even when only a small percentage of reviews are critical. The summaries may exaggerate negative themes, potentially harming sales. Amazon acknowledged the issue and said it is working to refine the technology based on seller feedback.
- Company involved
- Amazon
- AI system involved
- AI-powered review highlights
5 source articles · read the reporting →
AI used to finish painting that artist left incomplete
A social media post used AI to complete an unfinished painting, saying the story behind it was sad. Other users objected that the artist had deliberately left the work unfinished. One response said the artist's estate should sue.
7 source articles · read the reporting →
AI image generators produce misleading election images, study finds
A study by the Center for Countering Digital Hate found that leading AI image generators, including Midjourney, DreamStudio, ChatGPT Plus, and Microsoft Image Creator, could be manipulated to create misleading election-related images. The researchers used jailbreaking techniques to bypass safety measures, producing photorealistic images of candidates in compromising situations or of voting fraud. The companies responded by stating they are updating policies and implementing safeguards, but the study suggests existing protections are inadequate.
- Company involved
- Midjourney, Stability AI, OpenAI, Microsoft
- AI system involved
- Midjourney, DreamStudio, ChatGPT Plus, Microsoft Image Creator
8 source articles · read the reporting →
AI detectors falsely flag non-native English speakers' essays as AI-generated
A study by Stanford researchers found that seven popular AI text detectors wrongly flagged over half of essays written by non-native English speakers as AI-generated. The detectors assess text perplexity, and non-native speakers' simpler word choices lead to false positives. The researchers warn that this bias could have serious implications for students and job applicants, potentially leading to discrimination.
9 source articles · read the reporting →
AI script event in Tokyo canceled after plagiarism criticism
An event organizing company in Tokyo planned a performance where voice actors would read a script generated by ChatGPT, a generative AI. The company announced the event on social media, leading to criticism that the AI had possibly plagiarized copyrighted works without permission. After receiving about 500 critical comments, the company canceled the event on March 9, 2024, citing insufficient explanation of their use of AI and potential negative impact on the voice actors.
- AI system involved
- ChatGPT (paid subscription version)
3 source articles · read the reporting →
