🔍 Read the full analysis: What UBS Says About Its Staff’s AI Agents—and What It Doesn’t on ThorstenMeyerAI.com
Get everyday essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
UBS reported that employees are running more than 12,700 AI personal agents, alongside 608 AI applications in production and more than 1,090 in development. The bank has not disclosed what the agents can access or do, or how they are reviewed; CEO Sergio Ermotti said UBS does not allow outside models to act for clients.
UBS said on October 28 that employees are running more than 12,700 AI personal agents to support their daily work, offering a rare count of staff-built AI tools inside a major regulated bank. The third-quarter results also listed 608 AI applications in production and more than 1,090 in development, but did not explain what the agents can access, whether they can act independently or how the bank oversees them.
The figures appeared in UBS’s third-quarter results under a section on strategic investment. The bank reported nine large-scale AI initiatives, 608 live AI use cases—an 81% increase year over year—and more than 1,090 further applications in development. UBS also said more than 18,000 software developers use AI tools in their development cycles, describing the productivity gains as measurable without publishing a figure for them.
UBS said its US financial-adviser tool STAAT had generated more than 30 million insights so far in 2026. Reuters reported the agent count in its October 28 coverage of the results, and PYMNTS reported it the following day. Neither report, according to the source material, added operational detail beyond the bank’s disclosure.
The release did not define “personal agent.” UBS had said in its July 29 second-quarter results that all employees could create their own AI agents, describing the aim as helping staff spend more time on judgment, collaboration and client work. The bank’s technology materials also describe Red, an internal assistant available to about 100,000 employees, and an AI-powered human-resources assistant for tasks such as checking policies and booking leave. Those examples offer context, but UBS has not said they account for the 12,700 agents.
Infographic 01 · What UBS reported
12,700 personal AI agents, inside one bank
The AI figures from UBS’s third-quarter results, and how fast they grew in three months.
From permission to 12,700 agents
UBS 2Q26 release (29 Jul 2026) vs. 3Q26 release (28 Oct 2026)
What the agent number is not
- Not 12,700 people. One employee can run several agents; UBS does not say how many people use them.
- Not a definition. UBS does not say what an agent can access or do on its own.
- Not a result. “Measurable productivity gains”, but no hours or costs published.
- Scale, for context: about 13 agents per 100 internal full-time staff (12,700 ÷ 97,593, own calculation).
“Our employees are running over 12,700 AI personal agents to support them in their daily activities.”UBS, third-quarter 2026 results release, 28 October 2026
Also in the release: 9 large-scale AI initiatives and more than 30 million insights from the STAAT tool for US financial advisor teams so far this year. UBS placed 6th of 50 banks in the 2026 Evident AI Index.
Sources: UBS 3Q26 release, 28 Oct 2026 · UBS 2Q26 release, 29 Jul 2026. Reported identically by Reuters, 28 Oct 2026 and PYMNTS (29 Oct 2026).
Powered by Thorsten Meyer AI
The Oversight Question Behind the Count
The number shows that employee use of AI at UBS has moved beyond a general promise to adopt the technology: the bank says thousands of agents are already being run in day-to-day work. But a count alone does not show whether they save time, improve service or create new operational risks. UBS supplied no quantified productivity result for its developer tools and no breakdown of the agents’ tasks.
That gap matters in banking because an assistant that drafts a summary presents different risks from software that can send a message, change a record or initiate a transaction. Without details on permissions, human review and monitoring, readers cannot tell how much authority these agents have. The disclosure therefore gives a useful measure of adoption while leaving the more consequential questions about control unanswered.
The issue extends beyond UBS. The Swiss financial regulator FINMA expects banks to maintain a complete inventory of AI applications and assign risk classes. The source material says FINMA has noted that inventories are harder to maintain as applications become accessible to more people. UBS’s figure makes that governance challenge concrete, but does not establish that the bank has failed to meet regulatory expectations.
Products Worth Considering
LG UltraGear 27GX704A-B 27-inch QHD OLED Gaming Monitor
Samsung 27-inch Odyssey QD-OLED G8 4K 240Hz Gaming Monitor
Samsung 27 Odyssey OLED G5 (G50SF) QHD Gaming Monitor
As an affiliate, we earn on qualifying purchases.
From Employee Access to Reported Use
UBS’s October report followed its July statement that every employee had the ability to create an AI agent. The quarterly figures offer a later measure of reported use: 608 live applications, compared with 560 three months earlier, and more than 1,090 in development, up from more than 920. These are changes in counts reported by the bank, not evidence by themselves of how widely any particular tool is used or what outcomes it produces.
UBS reported that its workforce fell by about 1,500 full-time equivalents during the quarter, to 97,593. UBS and Reuters attributed the reduction to the Credit Suisse integration. The bank did not link the workforce change to AI, and the available reporting does not establish a connection in either direction.
UBS also cited its position in the 2026 Evident AI Index, which ranks 50 large banks. It said it placed sixth overall and was the only European bank in the top ten; JPMorganChase ranked first. These rankings are separate from the reported agent count and do not independently verify how the tools operate inside UBS.
“Employees are running over 12,700 AI personal agents to support them in their daily activities.”
— UBS, in its third-quarter results released October 28
What UBS Has Not Disclosed
UBS has not explained what qualifies as a “personal agent,” which models or platforms power them, what internal systems and data they can reach, or whether they can take actions without a person’s approval. It has also not said how many employees run agents, whether a single employee may run several, or what testing, logging and review processes apply.
The bank’s results do not identify how many of the 608 live AI use cases are customer-facing, nor do they specify the agents’ contribution to productivity or financial performance. Ermotti’s remarks draw a stated boundary around outside models acting for clients; they do not describe the permissions or safeguards for employee-run agents inside the bank. The available information does not establish whether those tools can initiate transactions.
These omissions are limits in the public disclosure, not proof that UBS lacks internal controls. The source material provides no inventory of controls, incidents or audit findings, and no response from UBS addressing those specific questions.
More Detail Needed on Controls
The next useful step would be for UBS to explain how it classifies and monitors employee-created agents: what information they may use, what actions they can take, when human approval is required and how the bank tracks performance and errors. The third-quarter results do not announce a timetable for such disclosures, so it is not clear when UBS will provide further detail.
For now, the bank’s reported counts can serve as a baseline for future updates, but they should not be treated as a measure of proven business impact or agent autonomy. Later results, public statements or regulatory information may clarify how UBS governs the tools. Until then, the confirmed picture is one of broad reported staff access and growing AI deployment, with the operating rules largely undisclosed.
Key Questions
How many AI agents did UBS report?
UBS said in its October 28, 2026, third-quarter results that employees were running more than 12,700 AI personal agents to support their daily activities.
What can UBS employees’ personal agents do?
UBS has not defined the term or published a task-by-task description. The bank’s separate materials describe tools for staff assistance, but the available reporting does not confirm which examples are included in the 12,700-agent figure.
Can AI agents at UBS act for clients?
CEO Sergio Ermotti said UBS does not allow outside AI models to enter the bank and act for clients, such as managing accounts or moving deposits. The statement does not detail what actions employee-run agents can take internally.
How many other AI applications did UBS report?
UBS listed 608 AI use cases in production, up 81% from a year earlier, and more than 1,090 applications in development. It also said more than 18,000 software developers use AI tools.
Did UBS link AI use to its staff reduction?
No. UBS and Reuters attributed the reduction of about 1,500 full-time equivalents in the quarter to the Credit Suisse integration. The bank did not link the workforce change to AI.
Source: ThorstenMeyerAI.com
