📌 TOPINDIATOURS Hot ai: Scientist Horrified as ChatGPT Deletes All His Research Wa

ChatGPT may be an excellent tool in case your strongly-worded email to your landlord about that ceiling leak needs a second pair of eyes. It also excels at coming up with a rough first draft for non-mission-critical writing, allowing you to carefully pick it apart and refine it.

But like all of its competitors, ChatGPT is plagued by plenty of well-documented shortcomings as well, from rampant hallucinations to a sycophantic tone that can easily lull users into gravely mistaken beliefs.

In other words, it’s not exactly a tool anybody should rely on to get important work done — and that’s a lesson University of Cologne professor of plant sciences Marcel Bucher learned the hard way.

In a column for Nature, Bucher admitted he’d “lost” two years’ worth of “carefully structured academic work” — including grant applications, publication revisions, lectures, and exams — after turning off ChatGPT’s “data consent” option.

He disabled the feature because he “wanted to see whether I would still have access to all of the model’s functions if I did not provide OpenAI with my data.”

But to his dismay, the chats disappeared without a trace in an instant.

“No warning appeared,” Bucher wrote. “There was no undo option. Just a blank page.”

The column was met with an outpouring of schadenfreude on social media, with users questioning how Bucher had gone two years without making any local backups. Others were enraged, calling on the university to fire him for relying so heavily on AI for academic work.

Some, however, did take pity.

“Well, kudos to Marcel Bucher for sharing a story about a deeply flawed workflow and a stupid mistake,” Heidelberg University teaching coordinator Roland Gromes wrote in a post on Bluesky. “A lot of academics believe they can see the pitfalls but all of us can be naive and run into this kind of problems!”

Bucher is the first to admit that ChatGPT can “produce seemingly confident but sometimes incorrect statements,” arguing that he never “equated its reliability with factual accuracy.” Nonetheless, he “relied on the continuity and apparent stability of the workspace,” using ChatGPT Plus as his “assistant every day.”

The use of generative AI in the scientific world has proven highly controversial.

Scientific journals are being flooded with poorly sourced AI slop, turning the process of peer review into a horror show, as The Atlantic reported this week. Entire fraudulent scientific journals are popping up to capitalize on others who are trying to get their AI slop published. The result? AI slop being peer-reviewed by AI models, further entrenching the polluting of scientific literature.

For their part, scientists are constantly being informed of how their work is being cited in various new papers — only to find that the referenced material was entirely hallucinated.

To be clear, there’s zero evidence that Bucher was in any way trying to sell off AI slop to his students or get dubious, AI-generated research published.

Nonetheless, his unfortunate experience with the platform should serve as a warning sign to others.

In his column, Bucher accused OpenAI of selling subscriptions to its ChatGPT Plus despite not assuring “basic protective measures” to stop years of his work from vanishing in an instant.

In a statement to Nature, OpenAI clarified that chats “cannot be recovered” after being deleted, and challenged Bucher’s claim that there was “no warning,” saying that “we do provide a confirmation prompt before a user permanently deletes a chat.”

The company also helpfully recommended that “users maintain personal backups for professional work.”

More on scientific AI slop: The More Scientists Work With AI, the Less They Trust It

The post Scientist Horrified as ChatGPT Deletes All His Research appeared first on Futurism.

🔗 Sumber: futurism.com

📌 TOPINDIATOURS Update ai: Which Agent Causes Task Failures and When?Researchers f

Share My Research is Synced’s column that welcomes scholars to share their own research breakthroughs with over 1.5M global AI enthusiasts. Beyond technological advances, Share My Research also calls for interesting stories behind the research and exciting research ideas. Contact us: chain.zhang@jiqizhixin.com

Meet the authors
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 failure-responsible agent and the decisive error step that led to the task’s failure.
2. 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 models, like OpenAI o1 and DeepSeek R1, find this task challenging.- This h…

Konten dipersingkat otomatis.

🔗 Sumber: syncedreview.com