(22 October 2025 – United States) Goldman Sachs is rolling out its “GS AI assistant” to 10,000 employees as part of the group’s plan to introduce “AI employees” as BofA seeks to avoid “sins of the past” with new forms of artificial intelligence (AI).
The internal GS AI platform supports drafting, summarising and code translation to speed-up repetitive work with plans to scale across the firm. GS executives states that the goal is an agentic coworker that behaves like a GS employee rather than a simple chatbot.
GS is also undertaking its “OneGS 3.0” initiative, targeting streamlined sales and client onboarding processes along with other critical areas such as lending, regulatory reporting and vendor management.
Critics highlight that AI models have been repeatedly shown to “hallucinate” facts, a significant issue that engineers are struggling to eliminate.
“The system should learn to check its own work and absorb the firm’s style over time. The stack is built with vendor models wrapped by internal guardrails and data layers to reduce leakage and keep outputs inside policy. Think about all the tasks that you might want to complete with regards to a variety of use cases for all those professions that can be now at your fingertips” commented Goldman Sachs CIO Marco Argenti.
“In a matter of three to five years, AI models could start to erode the lines between humans and AI. The AI assistant becomes really like talking to another GS employee. As we progress, the second step is when you’re starting to have this agentic behaviour, that is, ‘I’m completing a task on behalf of a Goldman employee, and I need to take a set of steps. That’s where the model is going to start to do things like a Goldman employee, not only say things like a Goldman employee” Argenti added.
Morgan Stanley released AskResearchGPT and Debrief to auto-summarise research and generate meeting notes. Analysts expect automation to reshape staffing, with Bloomberg Intelligence projecting up to 200,000 bank jobs at risk as routine tasks are absorbed by AI.
“We have a whole separate thought process about just grinding out stale and legacy applications. It takes time, and you have to get behind it and gradually pull them out of the ecosystem. But they’re almost two orthogonal problems embedded in your question” commented BofA Head of Platform and Head of Technology Ashok Krishnan and Duncan McInnes.