Artificial intelligence is no longer sitting quietly in the background of the modern workplace. AI assistants, meeting notetakers and agentic tools are now active contributors to daily business. However, the rush to deploy them is outpacing the frameworks needed to keep them in check.
Research from Theta Lake’s 7th Annual Digital Communications Governance Report, which surveyed 500 IT and compliance leaders, shows how big that gap has become. While 99% of financial services firms intend to expand their use of AI, 88% say they are already grappling with AI governance and data security challenges.
These problems are now a significant barrier to adoption. Theta Lake argues that firms which build governance in early, whether regulated or not, will be able to scale AI faster.
The first major shift is that AI will be treated as a participant in workplace communications rather than a simple productivity add-on.
As firms embed AI deeper into Unified Communications and Collaboration (UCC) platforms, a new category of interactions, dubbed “aiComms”, is set to grow sharply. AI is now drafting client emails, retrieving data on request and summarising client meetings.
Because these exchanges are conversational and evolve over time, reviewing isolated prompts and responses misses vital context. Any AI interaction linked to regulated activity will need to be captured, supervised and archived in the same way as a human’s.
Second, oversight will extend beyond AI outputs to how people use AI. Employees are increasingly trying to get around guardrails through techniques such as jailbreaking, while AI systems can still accidentally reveal personal data, client information, MNPI or confidential documents. Firms will need clear visibility of prompts, outputs and behaviours to spot data leaks, injection attempts and the use of unapproved tools.
Third, organisations are losing patience with “AI-washing”. Independent verification, particularly ISO/IEC 42001 certification for AI management systems, is expected to become a baseline requirement. The standard offers auditable evidence of responsible AI practices and aligns closely with the EU AI Act.
Fourth, regulators are making their position clear. FINRA’s 2026 Annual Regulatory Oversight Report includes a dedicated section on generative AI trends. It reiterates that firms remain accountable for communications whether a human or a machine created them.
The UK’s FCA has similarly confirmed that existing rules apply to AI-enabled activity. Firms will be expected to archive regulated AI communications, supervise AI outputs as rigorously as human content and control how AI tools are deployed. Accuracy and compliance of AI-generated content, already the top challenge cited by firms, is likely to attract closer scrutiny.
Finally, AI sprawl demands a unified strategy. With 82% of organisations using four or more UCC tools, and AI built into Microsoft Teams, Zoom, Webex and RingCentral, governing each tool separately risks blind spots. This is especially true as AI systems begin interacting with one another across platforms.
The firms that close these visibility gaps now will be best placed to capture AI’s full value while meeting regulatory expectations.
Read the full Theta Lake post here.
Copyright © 2026 FinTech Global
Investors
The following investor(s) were tagged in this article.
Leave a Reply