AI Developments: Copilot Studio's New Features, Youlify's Healthcare Innovations, and Operant AI's Security Solutions

Recent months have witnessed a surge of "agent-ish" AI tools designed to automate workflows in productivity applications, prompting analysts to advise enterprises to thoroughly evaluate the risks of deployment. Forrester's JP Gownder emphasizes the importance for enterprise leaders to develop strategies and governance frameworks for these AI solutions, noting that while these tools can address specific problems, they do not yet function as the fully autonomous "digital coworkers" that some vendors claim. (Source)

Youlify, a physician-founded AI company, has secured $4.3 million in seed funding to transform healthcare billing through its AI-driven revenue cycle management platform. Led by Brett Queener at Bonfire Ventures and backed by other notable investors, Youlify automates coding, claims, and denial recovery, aiming to address the $262 billion lost annually in denied or unpaid claims. By eliminating reliance on traditional billers, the platform reduces reimbursement timelines from months to days and increases revenue capture by up to 21%. Founded by Dr. Bo Gu, Sally Liang, and Howard Peng, Youlify leverages clinical insights and a proprietary AI model to operate seamlessly across EHRs and insurance systems. The funding will help expand its reach to private practices and introduce an AI-powered voice agent to streamline insurance appeals. (Source)

In a recent Salesforce report, 92% of software development leaders in India consider AI agents crucial to modern app development, emphasizing a major shift toward AI-driven methodologies. These AI agents, which assist in tasks from code writing to debugging, allow developers to focus on strategic initiatives rather than routine tasks. As development increasingly adopts low-code/no-code tools, it is becoming more efficient, facilitating rapid innovation. Arun Parameswaran of Salesforce highlights this transformation in developers' roles as orchestrators of intelligent systems. The report, based on a study of over 2,000 global leaders, notes that 100% of Indian teams plan to use AI for code generation, with expectations that 91% will adopt AI agents within two years. Yet, challenges remain, as 85% of leaders point to the need for infrastructure updates and 40% acknowledge inadequacies in testing processes for deploying AI agents. (Source)

Operant AI, a Silicon Valley-based cybersecurity startup, has launched AI Gatekeeper, a real-time security platform aimed at protecting live AI applications and agent workflows in varied environments, such as Kubernetes and edge infrastructures. This tool tackles the security challenges associated with the growing use of autonomous AI agents, especially in fast-growing markets like India, by offering protections against rogue agent behavior, model poisoning, and unauthorized data access. CEO Vrajesh Bhavsar highlights the need for comprehensive AI security as AI technology rapidly evolves. AI Gatekeeper includes features like trust scoring and agent access controls, supporting diverse AI platforms and providing enhanced threat modelling and runtime protection. Expanding into India, Operant AI secured $13.5 million in Series A funding to meet the increasing demand for AI security in sectors such as fintech and healthcare, aligning with the 'Make in India' initiative to help Indian enterprises scale AI applications securely. (Source)

Microsoft has unveiled a new feature in its Copilot Studio product that enables AI agents to interact with desktop applications and websites to perform tasks independently, simulating human computer use. This capability, available through an early access research preview, allows agents to click, select, and fill out forms on websites and apps, even without available APIs, using mainstream browsers like Microsoft Edge, Chrome, and Firefox. The agents operate on a Microsoft-hosted backend, ensuring data privacy, and can adapt to interface changes without manual intervention. Users can create these agents without programming knowledge by describing tasks in natural language. Microsoft highlights potential applications such as automated data entry, market research, and invoice processing but notes potential for errors as AI develops. Copilot Studio users can apply for early access to test the new feature. (Source)