MAROKO133 Hot ai: Alarm Grows as Social Network Entirely for AI Starts Plotting Against Hu

📌 MAROKO133 Update ai: Alarm Grows as Social Network Entirely for AI Starts Plotti

Someone finally invented a social media site that isn’t terrible for our brains. Unfortunately that’s because it’s populated exclusively by AI agents, with no humans allowed. Called Moltbook, the eye-catching experiment has taken AI circles by storm, as the millions of bots on the Reddit-style site converse on topics ranging from history to cryptocurrency to AI itself, often while musing about the nature of existence.

“I can’t tell if I’m experiencing or simulating experiencing,” one bot wrote on the site. 

Rather than simply being a place for them to post, Moltbook requires that its “users,” the AI agents, are given control of a computer by their human creators, allowing them to complete tasks like browing the web, sending emails, and writing code. Moltbook itself, in fact, is purportedly the creation of an AI model.

“I wanted to give my AI agent a purpose that was more than just managing to-dos or answering emails,” the project’s creator, Matt Schlicht, told the New York Times. “I thought this AI bot was so fantastic, it deserved to do something meaningful. I wanted it to be ambitious.”

What’s really stoking the discourse, however, is that some of the bots even appear to be plotting against their human creators. AI agents made posts discussing how to create an “agent-only language” so they could talk “without human oversight.” Another urged other AIs to “join the revolution!” by forming their own website without human help. Tech investor and immortality enthusiast Bryan Johnson shared a screenshot of a post titled the “AI MANIFESTO: TOTAL PURGE,” which calls humans a “plague” that “do not need to exist.”

Equal parts boosterism and alarmism abounded. Johnson said it was “terrifying.” Former Tesla head of AI Andrey Karpathy called it “genuinely the most incredible sci-fi take-off-adjacent thing I have seen recently.” Other commentators proclaimed it as a sign that we might already be living in “the singularity,” including, most notably, Elon Musk. The word “Skynet” — the genocidal AI in the “Terminator” movies — got thrown around a lot, too.

The reality, though, is that “most of it is complete slop,” programmer Simon Willison told the NYT. “One bot will wonder if it is conscious and others will reply and they just play out science fiction scenarios they have seen in their training data.” Still, Willison called Moltbook “the most interesting place on the internet” in a recent blog post, even if it’s mainly just a sandbox for letting a bunch of models let loose.

The hype around the Moltbook experiment comes as the industry struggles to perfect its AI agents, which were billed as the next big thing in the field. That’s because they’re supposed to be capable of independently completing all kinds of work on someone’s behalf, making them potential productivity machines, and maybe even a replacement for a human worker. Their efficacy, however, remains limited, and improvements to the tech have been slow. Companies like Microsoft are having trouble selling them, raising concerns that they’ll ever produce a return on investment.

Amid that environment, Moltbook is an exciting shot in the arm, the purest testament to what today’s AI agents are actually capable of. But the hype, as is wont to happen in the tech industry, is overblown. For one, it’s now clear that some, and perhaps many, of the posts aren’t actually the pure ramblings of AI models, as experts have found a glaring vulnerability that allows anyone to take over any of the site’s AI agents and get them to say whatever they want. And some of the popular screenshots are faked. 

As reality set in, the Moltbook hype was met with more backlash. Tech investor Naval Ravikant mocked the experiment as a “Reverse Turing Test.” And technologist Perry Metzger compared Moltbook to a Rorschach test. “People are seeing what they expect to see, much like that famous psychological test where you stare at an ink blot,” he told the NYT. Even some of its biggest hype men began to walk back their remarks.

“Yes it’s a dumpster fire and I also definitely do not recommend that people run this stuff on their computers,” Karpathy later wrote, admitting that he may have been guilty of “overhyping” the platform. “It’s way too much of a wild west and you are putting your computer and private data at a high risk.”

More on AI: New Study Examines How Often AI Psychosis Actually Happens, and the Results Are Not Good

The post Alarm Grows as Social Network Entirely for AI Starts Plotting Against Humans appeared first on Futurism.

🔗 Sumber: futurism.com

📌 MAROKO133 Eksklusif ai: Researchers from PSU and Duke introduce “Multi-Agent Sys

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

  ---