Dynamo AI

Yesterday, I had the privilege to speak to the House Financial Services Committee's AI Task Force, who invited Dynamo AI to describe our work productionizing some of the largest-scale deployments of GenAI across the financial sector. This historic session included an intimate discussion with Chairman of the U.S. House Committee on Financial Services Rep. Patrick McHenry, Vice Chairman Rep. French Hill, Rep. Sylvia Garcia, and Rep. Stephen Lynch. Also a part of the discussion was Daniel P. Huttenlocher, Dean of MIT Schwarzman College of Computing, and the Head of Massachusetts Institute of Technology EECS Department Asu Ozdaglar. It was also surreal to have this all take place right across from Dynamo's birthplace: CSAIL MIT! We had the chance to showcase to the task force emerging best-practices around AI redteaming, model controls in the context of SR 11-7, and regulatory guidance on AI. Moving forward, our continued collaboration between legislators, academia, and industry practitioners will be instrumental to ensure America maintains its competitive lead in both fostering AI leadership and reaping its advantages, while ensuring safe and secure adoption across highly regulated industries. I left the bipartisan session more energized than ever that our leaders on Capitol Hill are actively readying their teams to move swiftly on AI and deliver for the American people during this period of rapid change. Special thanks to Devina Khanna, Allison Behuniak, and Tammy Pham for making this possible!

We’re thrilled to announce Dynamo AI’s groundbreaking partnership with Qualcomm which leverages DynamoGuard to bring enterprise-grade security and compliance to Gen AI-powered notebooks. The Snapdragon X Elite platform is a critical step-function advancement in GenAI, enabling language models to run seamlessly on your notebook without needing to send any sensitive data or prompts to cloud LLMs. We’re proud to empower enterprises to safely and securely leverage on-device GenAI applications with DynamoGuard’s industry-leading, low-latency guardrails. Read more about Qualcomm’s coverage of our partnership here: https://hubs.la/Q02BpKfz0

What risk-mitigating controls may satisfy regulatory concerns surrounding third-party Generative AI use? Ever since the Federal Reserve Board, the Federal Deposit Insurance Corporation (FDIC), and the Office of the Comptroller of the Currency published joint guidance on third-party risk management, financial service risk and control experts have been debating effective strategies for oversight and ongoing monitoring of third-party Generative AI. While specific guidance to artificial intelligence was requested, the guidance was kept broad intentionally to allow for institutions to identify satisfactory control strategies. Dynamo AI's Daniel Ross, Head of Compliance Strategy, and Eric Lin, Head of Applied AI, provide insight into what they are hearing from financial service leadership teams and the strategies, processes, and evidence being considered. As financial service organizations continue to expand their use of third-party Generative AI, proven risk and control oversight and monitoring is crucial for these highly regulated entities. Enjoy the perspective! https://hubs.la/Q02zk9wt0

We knew GenAI compliance, red-teaming, and guardrailing was an important topic at Databricks Data + AI Summit 2024, but we didn't think it would be this hot of a ticket! https://lnkd.in/ezz_taBk Looking forward to seeing everyone in-person in San Francisco and for those who are unable to make it, we'll send out the recording from Eric Lin's fully booked session. Thank you to our partners at Databricks, Jefferson Ye, Josh Lillie, John Young, and Brian Dirking for the opportunity!

We had a blast last week as one of Microsoft's Spotlight Partners at #MSBuild showcasing DynamoEval (streamlined red-teaming), DynamoEnhance (remediate identified risks), and DynamoGuard (customizable multi-lingual guardrails) to support safe & secure Azure GenAI deployments for our joint customers and developers. Can't wait to showcase our new Copilot+ PC & DynamoGuard modules along with our friends at Lenovo 😏 Thank you for the opportunity Microsoft for Startups, Shelby Swofford, Aruna Dadi, Adam Higgs, and Erica Cangelosi!

Last week, Dynamo's Chief Product Officer, Christian Lau, announced a suite of exciting Dynamo product launches and case studies of productionizing GenAI to a room of AI leaders from leading financial institutions and fintechs at Finovate Spring. We're excited to define a path for financial services to deliver high ROI Generative AI, while helping our enterprise customers to navigate the emerging guidance and requirements from regulators.

The Dynamo team had a fantastic set of meetings in DC with key policy makers shaping the future of AI laws and regulations. It was amazing to witness these policy makers' enthusiasm for both spurring AI innovation, while enabling safe and responsible AI adoption. At Dynamo we've spoken to 100+ legislators, government agencies, and regulators to both educate policy makers on Generative AI development in highly regulated sectors and align our Dynamo products with emerging requirements from governments. Our team is committed to ensuring strong collaboration between both the public and private sectors to get this right.

Thank you Lenovo for the opportunity to partner together in ensuring safe, secure, and efficient GenAI deployments for our customers. We are super excited to showcase our joint Copilot+ PC offerings, optimized for Lenovo devices - more news to come! https://hubs.la/Q02ybp970

Looking forward to leading the discussion on Thursday (5/23 at 4PM EST) with the one and only Martin Stanley, CISSP, who leads AI strategy for the Department of Homeland Security Cybersecurity and Infrastructure Security Agency (CISA) and is now driving adoption of the AI Risk Management Framework for the National Institute of Standards and Technology (NIST). For many of my financial services colleagues, insights into how they can best adopt the NIST AI RMF is critical, and judging by the questions received, this should be a great discussion. Please register for this Dynamo AI event here: https://lnkd.in/e43Se3mM

🛡DynamoGuard: A 10X faster and more performant approach to moderating harmful content ⚖ DynamoGuard outperforms today’s content moderation solutions on detecting harmful content, while operating at 10x lower latency and providing essential explainability capabilities to end users and compliance teams. 📶 To achieve these gains, DynamoGuard presents an enterprise-ready guardrail that adapts to the subtleties and nuances that exist in noisy real-world user conversations, driven by not only Dynamo's own proprietary industry-leading safety optimizations but also with a platform for integrating human feedback from key enterprise stakeholders like legal, compliance, infosec, and product teams to address gaps. Read more from DynamoGuard Product Manager Shaily Fozdar, research scientist James O'Neill, and our Head of Applied AI Eric Lin in our post below: https://lnkd.in/gmRjxirW