Tag
This paper analyzes why enterprise AI deployments stall in regulated firms, proposing a production bar that includes accuracy, reproducibility, groundedness, and detectability. It measures the human review burden across model and tool configurations, showing that confidence signals and source citation can cut review from 100% to 49% but self-verification adds latency without improving error tolerance.
This paper presents a methodology for creating high-fidelity synthetic customer agents (digital twins) grounded in real transactional and conversational data for large-scale validation of LLM-based chatbots in regulated domains like banking.
Samantha McBride's company migrated from a legacy stack to MongoDB to build a flexible financial services platform, enabling easier AI agent deployment without client data restructuring.
ServiceNow invests $40 million in Indian banking software specialist BusinessNext at a $700 million valuation, deepening its push into AI-driven financial services through a strategic partnership.
Gergely Orosz reflects on areas where AI has not accelerated progress, such as launching financial services across regions and supporting old hardware, noting that software was never the main bottleneck.
The UK's Financial Conduct Authority warns of an 'arms race' to keep up with AI use in financial services, recommending new powers to supervise critical third-party tech providers and proposing a free AI-enabled financial guidance service.
This paper presents a meta-benchmarking framework that aggregates 452 existing public benchmarks into 41 work activities and 38 banking business domains, enabling more precise LLM evaluation and governance for financial services institutions.
This open-source project adapts Anthropic's financial services skill pack for China's A-share market, offering 63 Claude Skills with support for data sources like Wind and iFind, helping financial professionals efficiently complete tasks such as research reports and valuations.
LSEG partners with OpenAI to deploy ChatGPT Enterprise and APIs across its organization, reducing product release cycles from months to weeks and enabling thousands of employees globally.
OQC, JPMorganChase, and AMD have launched a research collaboration to develop a dedicated Quantum-AI Data Centre in London, integrating OQC's GENESIS quantum system with AMD-supported AI and classical computing for financial services applications.
Financial institutions are shifting from siloed AI models to unified transaction foundation models built on transformer architectures, as demonstrated by NVIDIA's report and Revolut's PRAGMA model, which improves fraud detection, credit scoring, and recommendations while reducing feature engineering effort.
MUFG, one of Japan's largest financial groups, is partnering with OpenAI to roll out ChatGPT Enterprise to approximately 35,000 employees, aiming to become an AI-native company by integrating AI into daily workflows and customer experiences.
Anthropic has released a GitHub repository for financial services workflows, including agent workflows for investment banking, equity research, private equity, and wealth management. Users can deploy via the Claude Cowork plugin or Managed Agents API, and by entering simple instructions, it automatically generates professional investment research reports, financial analysis, and more.
This paper introduces five governance metrics to quantify policy compliance at the decision rationale level for LLMs in regulated financial workflows, finding that mechanical enforcement (operating outside the model's interpretive loop) reduces non-informative deferrals by 73% and reveals governance-task decoupling: text-only governance degrades on both dimensions under stress, while mechanical enforcement preserves governance quality even as task performance drops.
The article discusses how financial services companies must ensure data quality, security, and accessibility to successfully deploy agentic AI, emphasizing that the technology's effectiveness depends more on robust data foundations than on system sophistication.
The article reviews the top five AI Agent projects on GitHub with the fastest star growth recently, highlighting Anthropic's Financial Services Agent Framework, ByteDance's UI-TARS Desktop, and various coding Agent tools.
Anthropic has open-sourced the financial-services repository, offering 9 out-of-the-box AI agents for investment banking and 11 MCP data connectors, aiming to automate workflows such as financial modeling, research report writing, and compliance reviews.
SecureLend Agents is a new product launching AI-driven underwriting agents designed for venture capitalists, lenders, and insurers to automate risk assessment.
Singular Bank has deployed an internal AI assistant called Singularity, powered by ChatGPT and Codex, to help bankers analyze portfolios and draft communications, saving 60-90 minutes per day.
Anthropic releases agent templates for financial services tasks such as pitches, KYC, and closing books, available via Claude Code.