MAROKO133 Breaking ai: Anthropic rolls out Claude AI for finance, integrates with Excel to

📌 MAROKO133 Update ai: Anthropic rolls out Claude AI for finance, integrates with

Anthropic is making its most aggressive push yet into the trillion-dollar financial services industry, unveiling a suite of tools that embed its Claude AI assistant directly into Microsoft Excel and connect it to real-time market data from some of the world's most influential financial information providers.

The San Francisco-based AI startup announced Monday it is releasing Claude for Excel, allowing financial analysts to interact with the AI system directly within their spreadsheets — the quintessential tool of modern finance. Beyond Excel, select Claude models are also being made available in Microsoft Copilot Studio and Researcher agent, expanding the integration across Microsoft's enterprise AI ecosystem. The integration marks a significant escalation in Anthropic's campaign to position itself as the AI platform of choice for banks, asset managers, and insurance companies, markets where precision and regulatory compliance matter far more than creative flair.

The expansion comes just three months after Anthropic launched its Financial Analysis Solution in July, and it signals the company's determination to capture market share in an industry projected to spend $97 billion on AI by 2027, up from $35 billion in 2023.

More importantly, it positions Anthropic to compete directly with Microsoft — ironically, its partner in this Excel integration — which has its own Copilot AI assistant embedded across its Office suite, and with OpenAI, which counts Microsoft as its largest investor.

Why Excel has become the new battleground for AI in finance

The decision to build directly into Excel is hardly accidental. Excel remains the lingua franca of finance, the digital workspace where analysts spend countless hours constructing financial models, running valuations, and stress-testing assumptions. By embedding Claude into this environment, Anthropic is meeting financial professionals exactly where they work rather than asking them to toggle between applications.

Claude for Excel allows users to work with the AI in a sidebar where it can read, analyze, modify, and create new Excel workbooks while providing full transparency about the actions it takes by tracking and explaining changes and letting users navigate directly to referenced cells.

This transparency feature addresses one of the most persistent anxieties around AI in finance: the "black box" problem. When billions of dollars ride on a financial model's output, analysts need to understand not just the answer but how the AI arrived at it. By showing its work at the cell level, Anthropic is attempting to build the trust necessary for widespread adoption in an industry where careers and fortunes can turn on a misplaced decimal point.

The technical implementation is sophisticated. Claude can discuss how spreadsheets work, modify them while preserving formula dependencies — a notoriously complex task — debug cell formulas, populate templates with new data, or build entirely new spreadsheets from scratch. This isn't merely a chatbot that answers questions about your data; it's a collaborative tool that can actively manipulate the models that drive investment decisions worth trillions of dollars.

How Anthropic is building data moats around its financial AI platform

Perhaps more significant than the Excel integration is Anthropic's expansion of its connector ecosystem, which now links Claude to live market data and proprietary research from financial information giants. The company added six major new data partnerships spanning the entire spectrum of financial information that professional investors rely upon.

Aiera now provides Claude with real-time earnings call transcripts and summaries of investor events like shareholder meetings, presentations, and conferences. The Aiera connector also enables a data feed from Third Bridge, which gives Claude access to a library of insights interviews, company intelligence, and industry analysis from experts and former executives. Chronograph gives private equity investors operational and financial information for portfolio monitoring and conducting due diligence, including performance metrics, valuations, and fund-level data.

Egnyte enables Claude to securely search permitted data for internal data rooms, investment documents, and approved financial models while maintaining governed access controls. LSEG, the London Stock Exchange Group, connects Claude to live market data including fixed income pricing, equities, foreign exchange rates, macroeconomic indicators, and analysts' estimates of other important financial metrics. Moody's provides access to proprietary credit ratings, research, and company data covering ownership, financials, and news on more than 600 million public and private companies, supporting work and research in compliance, credit analysis, and business development. MT Newswires provides Claude with access to the latest global multi-asset class news on financial markets and economies.

These partnerships amount to a land grab for the informational infrastructure that powers modern finance. Previously announced in July, Anthropic had already secured integrations with S&P Capital IQ, Daloopa, Morningstar, FactSet, PitchBook, Snowflake, and Databricks. Together, these connectors give Claude access to virtually every category of financial data an analyst might need: fundamental company data, market prices, credit assessments, private company intelligence, alternative data, and breaking news.

This matters because the quality of AI outputs depends entirely on the quality of inputs. Generic large language models trained on public internet data simply cannot compete with systems that have direct pipelines to Bloomberg-quality financial information. By securing these partnerships, Anthropic is building moats around its financial services offering that competitors will find difficult to replicate.

The strategic calculus he…

Konten dipersingkat otomatis.

🔗 Sumber: venturebeat.com

📌 MAROKO133 Eksklusif ai: Which Agent Causes Task Failures and When?Researchers fr

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: [email protected]

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