MUFG builds internally compliant version of Claude Code
Japanese bank says Anthropic’s coding and productivity tool is effective at obtaining information but bucks compliance guardrails, prompting the creation of an in-house alternative.
Mitsubishi UFJ Financial Group (MUFG) has designed a proprietary version of Anthropic’s Claude Code AI harness because the coding tool was too effective at accessing information and bypassing the bank’s security and compliance guidelines.
Stu Taylor, the bank’s new head of global markets AI for all regions except Japan, tells WatersTechnology that while the AI tools the bank utilizes are all human-in-the-loop, public agentic AI “harnesses”—tools like Claude Code, Cursor, and GitHub Copilot
Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.
To access these options, along with all other subscription benefits, please contact info@waterstechnology.com or view our subscription options here: https://subscriptions.waterstechnology.com/subscribe
You are currently unable to print this content. Please contact info@waterstechnology.com to find out more.
You are currently unable to copy this content. Please contact info@waterstechnology.com to find out more.
Copyright Infopro Digital Limited. All rights reserved.
As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (point 2.4), printing is limited to a single copy.
If you would like to purchase additional rights please email info@waterstechnology.com
Copyright Infopro Digital Limited. All rights reserved.
You may share this content using our article tools. As outlined in our terms and conditions, https://www.infopro-digital.com/terms-and-conditions/subscriptions/ (clause 2.4), an Authorised User may only make one copy of the materials for their own personal use. You must also comply with the restrictions in clause 2.5.
If you would like to purchase additional rights please email info@waterstechnology.com
More on Emerging Technologies
Few banks formally evaluate GenAI human-in-the-loop controls
Benchmarking: G-Sibs and challengers use tools to test controls efficacy; others rely on judgment.
New Preqin indices, Bloomberg’s TCA API, and more
The Waters Cooler: A recap of the major tech and data news from the past week in the capital markets.
Agentic AI for XVAs still five years away, experts say
Risk Live: Questions about validation and accountability keep autonomous AI models out of XVAs for now.
Finance firms join Anthropic’s Project Glasswing as AI transforms cyber risk
Broadridge and Intercontinental Exchange are taking part in Anthropic’s cybersecurity initiative, built around its controversial Mythos model.
NSCC extends hours, Tradeweb finds on-chain success, and more
The Waters Cooler: A recap of the major tech and data news from the past week in the capital markets.
Waters Wavelength Podcast Ep. 355: Faculty’s David Bholat
This week, David Bholat joins the podcast to talk about AI deployment.
Managing AI models is reshaping three lines of defense, say banks
Model managers want a seat at table during development, and expect the first line to take charge of AI model testing.
Trumid and BlackRock partner, reference data drama, Finos launches AI fund, and more
The Waters Cooler: A recap of the major tech and data news from the past week in the capital markets.
