MAROKO133 Breaking ai: Woman Suffers AI Psychosis After Obsessively Generating AI Images o

📌 MAROKO133 Update ai: Woman Suffers AI Psychosis After Obsessively Generating AI

On top of the environmental, political, and social toll AI has taken on the world, it’s also been linked to a severe mental health crisis in which users are spiraling into delusions and ending up committed to psychiatric institutions, or even dead by suicide.

Take Caitlin Ner. Writing in an essay for Newsweek, Ner discusses her experience as head of user experience at an AI image generator startup — a gig she says pulled her into the throes of an AI-induced mental health breakdown.

In her tell-all, Ner says it all began on the job, where she spent upward of nine hours a day prompting early, 2023-era generative AI systems. Though the faux-human images it spat out were often mangled and twisted, it still “felt like magic” — at least at first.

“Within a few months, that magic turned manic,” she wrote.

Ner wrote that these early images “started to distort my body perception and overstimulate my brain in ways that were genuinely harmful to my mental health.” Yet even when the AI learned to take it easy on the number of fingers it generated on a human hand, its images still took a mental toll, trading anatomical errors for scenes inhabited by impossibly slim, beautiful figures.

“Seeing AI images like this over and over again rewired my sense of normal,” Ner explained. “When I’d look at my real reflection, I’d see something that needed correction.”

At one pivotal moment, Ner began experimenting with AI images depicting herself as a fashion model, a directive set down by her company, which was pursuing users interested in fashion. “I caught myself thinking, ‘if only I looked like my AI version,’” she wrote. “I was obsessed with becoming skinnier, having a better body and perfect skin.”

She soon began losing sleep in order to generate more and more images, which she called “addictive,” because each image triggered a “small burst of dopamine.” Though Ner had been successfully treating her bipolar disorder prior to her foray into AI fashion modeling, this new obsession spun into a “manic bipolar episode,” which she says triggered an episode of psychosis.

“When I saw an AI-generated image of me on a flying horse, I started to believe I could actually fly,” Ner writes. “The voices told me to fly off my balcony, made me feel confident that I could survive. This grandiose delusion almost pushed me to actually jump.”

Luckily, she caught herself and began reaching out to friends and family for help. A clinician helped her realize her work had triggered the spiral, leading her to leave the AI startup. “I now understand that what happened to me wasn’t just a coincidence of mental illness and technology,” she explains. “It was a form of digital addiction from months and months of AI image generation.”

She has since jumped ship to become a director at another trendy company, PsyMed Ventures, which Newsweek described as a VC fund investing in mental and brain health. Many of the companies PsyMed invests in feature AI tools — which Ner says she still uses, albeit with a newfound sense of respect.

More on AI: Man Describes How ChatGPT Led Him Straight Into Psychosis

The post Woman Suffers AI Psychosis After Obsessively Generating AI Images of Herself appeared first on Futurism.

🔗 Sumber: futurism.com

📌 MAROKO133 Update ai: Researchers from PSU and Duke introduce “Multi-Agent System

Share My Research is Synced’s column that welcomes scholars to share their own research breakthroughs with over 2M global AI enthusiasts. Beyond technological advances, Share My Research also calls for interesting stories behind the research and exciting research ideas. 

Meet the author
Institutions: Penn State University, Duke University, Google DeepMind, University of Washington, Meta, Nanyang Technological University, and Oregon State University. The co-first authors are Shaokun Zhang of Penn State University and Ming Yin of Duke University.

In recent years, LLM Multi-Agent systems have garnered widespread attention for their collaborative approach to solving complex problems. However, it’s a common scenario for these systems to fail at a task despite a flurry of activity. This leaves developers with a critical question: which agent, at what point, was responsible for the failure? Sifting through vast interaction logs to pinpoint the root cause feels like finding a needle in a haystack—a time-consuming and labor-intensive effort.
 
This is a familiar frustration for developers. In increasingly complex Multi-Agent systems, failures are not only common but also incredibly difficult to diagnose due to the autonomous nature of agent collaboration and long information chains. Without a way to quickly identify the source of a failure, system iteration and optimization grind to a halt.
 
To address this challenge, researchers from Penn State University and Duke University, in collaboration with institutions including Google DeepMind, have introduced the novel research problem of “Automated Failure Attribution.” They have constructed the first benchmark dataset for this task, Who&When, and have developed and evaluated several automated attribution methods. This work not only highlights the complexity of the task but also paves a new path toward enhancing the reliability of LLM Multi-Agent systems.
The paper has been accepted as a Spotlight presentation at the top-tier machine learning conference, ICML 2025, and the code and dataset are now fully open-source.

Paper：https://arxiv.org/pdf/2505.00212
Code：https://github.com/mingyin1/Agents_Failure_Attribution
Dataset：https://huggingface.co/datasets/Kevin355/Who_and_When
 
 
Research Background and Challenges
LLM-driven Multi-Agent systems have demonstrated immense potential across many domains. However, these systems are fragile; errors by a single agent, misunderstandings between agents, or mistakes in information transmission can lead to the failure of the entire task.

Currently, when a system fails, developers are often left with manual and inefficient methods for debugging:
Manual Log Archaeology : Developers must manually review lengthy interaction logs to find the source of the problem.
Reliance on Expertise : The debugging process is highly dependent on the developer’s deep understanding of the system and the task at hand.
 
This “needle in a haystack” approach to debugging is not only inefficient but also severely hinders rapid system iteration and the improvement of system reliability. There is an urgent need for an automated, systematic method to pinpoint the cause of failures, effectively bridging the gap between “evaluation results” and “system improvement.”

Core Contributions
This paper makes several groundbreaking contributions to address the challenges above:
1. Defining a New Problem: The paper is the first to formalize “automated failure attribution” as a specific research task. This task is defined by identifying the

2. failure-responsible agent and the decisive error step that led to the task’s failure.

Constructing the First Benchmark Dataset: Who&When : This dataset includes a wide range of failure logs collected from 127 LLM Multi-Agent systems, which were either algorithmically generated or hand-crafted by experts to ensure realism and diversity. Each failure log is accompanied by fine-grained human annotations for:
Who: The agent responsible for the failure.
When: The specific interaction step where the decisive error occurred.
Why: A natural language explanation of the cause of the failure.

3. Exploring Initial “Automated Attribution” Methods : Using the Who&When dataset, the paper designs and assesses three distinct methods for automated failure attribution:
All-at-Once: This method provides the LLM with the user query and the complete failure log, asking it to identify the responsible agent and the decisive error step in a single pass. While cost-effective, it may struggle to pinpoint precise errors in long contexts.
Step-by-Step: This approach mimics manual debugging by having the LLM review the interaction log sequentially, making a judgment at each step until the error is found. It is more precise at locating the error step but incurs higher costs and risks accumulating errors.
Binary Search: A compromise between the first two methods, this strategy repeatedly divides the log in half, using the LLM to determine which segment contains the error. It then recursively searches the identified segment, offering a balance of cost and performance.
 
Experimental Results and Key Findings

Experiments were conducted in two settings: one where the LLM knows the ground truth answer to the problem the Multi-Agent system is trying to solve (With Ground Truth) and one where it does not (Without Ground Truth). The primary model used was GPT-4o, though other models were also tested. The systematic evaluation of these methods on the Who&When dataset yielded several important insights:

• A Long Way to Go: Current methods are far from perfect. Even the best-performing single method achieved an accuracy of only about 53.5% in identifying the responsible agent and a mere 14.2% in pinpointing the exact error step. Some methods performed even worse than random guessing, underscoring the difficulty of the task.
• No “All-in-One” Solution: Different methods excel at different aspects of the problem. The All-at-Once method is better at identifying “Who,” while the Step-by-Step method is more effective at determining “When.” The Binary Search method provides a middle-ground performance.

• Hybrid Approaches Show Promise but at a High Cost: The researchers found that combining different methods, such as using the All-at-Once approach to identify a potential agent and then applying the Step-by-Step method to find the error, can improve overall performance. However, this comes with a significant increase in computational cost.

• State-of-the-Art Models Struggle: Surprisingly, even the most advanced reasoning m…

  Konten dipersingkat otomatis.

  🔗 Sumber: syncedreview.com

  ---