Navigating Compliance Challenges for Financial Institutions with AI Solutions
Financial institutions are encountering rising costs in regulatory compliance. A recent survey reveals that 76% of financial services companies have increased their compliance spending between 2022 and 2023, largely attributing this rise to new legislative requirements. With compliance expenses averaging around $10,000 per employee, many firms are actively seeking strategies to cut costs while ensuring adherence to regulations. Entrepreneurs Nir Laznik and Eyal Peleg believe they have an effective solution, leveraging the capabilities of generative AI—a growing trend in the industry.
Laznik and Peleg co-founded Sedric, an AI-driven platform tailored to assist financial institutions in implementing compliance rules and identifying potential issues. Before founding Sedric, Laznik led multiple startups, most notably in developing photo kiosk software, while Peleg dedicated nearly eight years to AI and machine learning at Intel.
"We recognized the disproportionate pressure on mid-sized organizations and the evolving challenges faced by banks," Laznik explained. "The swift progress in AI technology opened new pathways to address these issues. This confluence of challenges and advancements inspired the creation of Sedric."
Sedric's AI functions as a vigilant overseer, monitoring various communication channels such as calls, chats, emails, social media messages, and instant communications. It seeks to flag compliance issues—including missed disclosures, procedural oversights, and misconduct—as they occur. Additionally, Sedric can automatically address problems and offer coaching to staff involved, according to Laznik.
"This technology equips compliance officers with a comprehensive view of customer interactions across diverse channels, enabling swift identification of any deviations from compliance policies," Laznik remarked. "Our platform spans the entire compliance lifecycle, encompassing policy formulation, enforcement, correction, and auditing."
The in-depth monitoring capabilities of Sedric may raise concerns about intrusion, particularly as it assigns compliance scores to employees based on policy adherence. However, existing U.S. state and federal regulations grant businesses significant leeway in monitoring staff, provided they maintain transparency in their practices.
Furthermore, certain federal regulations, particularly those concerning insider trading, collusion, and sharing sensitive earnings documents, require financial institutions to diligently track employee interactions with customers and the market. These regulations take precedence over state laws, like those in New York and Connecticut, which impose additional scrutiny on workforce monitoring.
In discussing the potential for bias in Sedric’s AI—given its role in monitoring staff from diverse backgrounds—I raised concerns about discriminatory practices that can result from biased AI models. Past studies have shown that some AI trained to identify negative behavior may misinterpret phrases in African-American Vernacular English as "toxic," and that speech recognition tools often struggle with accuracy when transcribing audio from Black speakers compared to their white counterparts.
Laznik assured me that Sedric employs "fine-tuned models" developed using proprietary datasets curated in collaboration with industry specialists to minimize bias. The company also continuously monitors the performance of deployed AI models and retrains them as needed.
"Our platform allows clients to provide direct feedback through various annotation inputs, which compliance teams vet and use for retraining or integrating into the prediction processes,” he elaborated. “This ensures our models are increasingly customized to meet each client's unique needs."
To safeguard both customer and employee privacy, Sedric offers companies the flexibility to determine data storage locations and implement controls that attempt to redact personally identifiable information.
“At Sedric, we've built our platform with compliance and security as core priorities,” Laznik stated. “Enterprises are empowered to establish their own data retention policies and compliance standards per their internal guidelines and relevant regulations.”
Currently, Sedric provides tools to support call center agents during client interactions and boasts a client base of "hundreds" of compliance officers and enterprises across the U.S. and Europe, according to Laznik. Revenue has reportedly surged fivefold over the past year, although specific figures were not disclosed.
“For small and medium-sized businesses, we provide an off-the-shelf solution, while enterprises and banks benefit from a hybrid model with customized features,” Laznik highlighted. “Our focus on the specific needs of financial institutions, combined with our exclusive library of pre-trained, regulation-inspired models, uniquely positions us in the market.”
Focusing specifically on financial sector clients has proven advantageous for Sedric, distinguishing it from competitors in the crowded and often contentious workplace monitoring space, including Fairwords, Shield, Erudit, and Aware. Despite the competitive landscape, investor interest remains strong, particularly as AI technology becomes more intertwined with compliance solutions.
In a noteworthy development, Foundation Capital has led an $18.5 million Series A funding round for Sedric, with participation from Amex Ventures. This new funding aims to significantly expand the company’s go-to-market strategies and R&D teams in New York City and Tel Aviv, raising Sedric’s total funding to $22 million.
Looking ahead, Sedric plans to double its workforce within the next year.
Ranking
