📌 TOPINDIATOURS Update ai: Grab Your Betrayal-Themed Popcorn Buckets, Because Micr

Mutual success has only heightened tensions between Microsoft and OpenAI, whose partnership helped kickstart the AI boom. Now, the Financial Times reports the latest escalation between the two heavyweights: Microsoft is considering suing OpenAI.

The drama is a bit of a corporate love triangle, and concerns a $50 billion deal to offer OpenAI’s new product, Frontier, on Amazon Web Services, the e-commerce giant’s cloud computing platform. Microsoft believes this violates the spirit and the letter of its own exclusivity agreement with OpenAI, which holds that all access to OpenAI’s AI models should only be through Microsoft’s Azure cloud platform.

Frontier is still unreleased, and no legal action has been set into motion yet, as both companies are still in talks. But Microsoft is signaling that it isn’t gun-shy.

“We know our contract,” a person familiar with Microsoft’s position told the FT. “We will sue them if they breach it. If Amazon and OpenAI want to take a bet on the creativity of their contractual lawyers, I would back us, not them.”

Once, OpenAI depended on Microsoft’s billions of dollars of investment. But with the runaway success of ChatGPT, it’s now approaching a trillion dollar valuation and has long started seeking greater independence from its patron, which isn’t quite willing to let go, especially as OpenAI has become one of its biggest competitors. 

Last year, the two clashed as OpenAI restructured itself into a for-profit public benefit corporation, but eventually settled on a new agreement in September. Microsoft relinquished its right to be OpenAI’s exclusive cloud provider, but retained a clause that requires all of OpenAI’s API calls, or calls to access its AI models, to be routed through Azure, the reporting noted.

Microsoft now feels that OpenAI is trying to weasel its way around this clause, with both companies’ lawyers fighting for weeks over the latter’s agreement with Amazon, sources told the FT. 

With Frontier, Amazon and OpenAI are developing a “Stateful Runtime Environment” that runs in Amazon’s Bedrock AI platform, designed to access data stored on AWS and allow OpenAI agents to keep context, remember prior work, integrate across different software tools, and access computing power, in effect making it more useful for ongoing projects. 

But Microsoft doesn’t believe it’s possible for OpenAI to let Amazon run its AI tech without violating the API clause. Amazon is seemingly aware of how suspicious this looks, with an internal memo providing strict instructions to staff to describe the SRE as only being “powered by,” “enabled by” or “integrates with” OpenAI, but not phrases like “enables access” to OpenAI’s tech, per the FT.

A lawsuit, however, would be inconvenient for both parties. One source told the FT that Microsoft was unlikely to sue OpenAI because it would invite further scrutiny while it’s already facing investigations in the US, UK, and EU into its alleged anti-competitive licensing practices with Azure.

Meanwhile, OpenAI is reportedly gunning to go public in a historic trillion dollar IPO. A lawsuit could throw a wrench into those plans, which are already being hamstrung by a lawsuit by Elon Musk accusing OpenAI of abandoning its beneficent non-profit roots.

“The last thing OpenAI needs is another court case right now,” the person familiar with Microsoft’s position told the FT.

More on AI: Panicked OpenAI Execs Cutting Projects as Walls Close In

The post Grab Your Betrayal-Themed Popcorn Buckets, Because Microsoft Is Threatening to Sue OpenAI appeared first on Futurism.

🔗 Sumber: futurism.com

📌 TOPINDIATOURS Update ai: Researchers from PSU and Duke introduce “Multi-Agent Sy

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

  ---