The BNP Paribas AI tool marks a clear step toward embedding artificial intelligence into the everyday workflows of investment banking. Rather than experimenting at the edges, the bank is testing how AI can support one of the most time-consuming core activities for bankers: preparing client pitch materials under tight deadlines.
According to Financial News, BNP Paribas has rolled out an internal system known as IB Portal. The tool is designed to help bankers assemble pitch decks faster by reusing existing internal materials instead of recreating content from scratch. In investment banking, pitch preparation often involves repeated searches for similar slides, charts, and deal precedents already produced elsewhere in the firm.
This repetition is costly in both time and focus. Teams frequently rebuild the same market views and transaction analyses, even when comparable work exists across regions or business lines. The BNP Paribas AI tool aims to reduce this inefficiency by acting as an intelligent internal search engine for pitch content.
IB Portal scans the bank’s historical pitch materials and uses what BNP Paribas describes as “smart prompts” to surface relevant slides, analyses, and supporting data for a new mandate. Instead of starting with a blank deck, bankers receive curated internal content that can be adapted to the specific client and situation.
George Holst, head of the corporate clients group at BNP Paribas, explained that the tool helps teams find what matters most ahead of meetings and pitches. He noted that it can reduce research time by days, allowing bankers to focus more on strategic thinking, judgement, and client relationships rather than repetitive preparation work.
What makes this deployment notable is its focus on real, constrained workflows. Pitch decks are not generic documents. They contain internal viewpoints, sensitive client details, and information subject to regulatory oversight. As a result, making an AI system useful in this environment depends heavily on structure rather than conversational flexibility.
To function safely, the BNP Paribas AI tool must respect strict access controls across regions and business lines. Not all materials are visible to every user, and content must move through clear stages before becoming client-ready. Traceability is also critical. Bankers need to know where information comes from and how it has been used previously.
Human oversight remains central. Any output generated or retrieved by IB Portal still requires review before it leaves the firm. Without these checks, the risks of errors, hallucinated data, or inappropriate disclosure increase significantly, especially in a regulated industry like banking.
IB Portal also fits into a wider AI strategy at BNP Paribas. In June 2025, the bank announced an internal “LLM as a Service” platform. This initiative provides business units with shared access to large language models hosted within BNP Paribas data centres, supported by dedicated GPU infrastructure.
The platform is managed by internal IT teams and supports a mix of models, including open-source systems and models from Mistral AI. The bank has also indicated plans to deploy models trained on its own internal data. Target use cases include document drafting, information retrieval, and internal AI assistants across different functions.
BNP Paribas is not alone in this approach. Other global banks are investing heavily in controlled, in-house AI platforms. JPMorganChase has highlighted its internal “LLM Suite,” while Goldman Sachs has rolled out a proprietary GS AI Assistant. UBS has discussed using an internal M&A co-pilot, and specialist tools such as Rogo are gaining traction at firms like Nomura and Moelis.
For BNP Paribas, the real measure of success will be adoption. The benefits are clear: less time spent searching, fewer duplicated decks, and better reuse of institutional knowledge. At the same time, the risks are well understood, including data accuracy, source clarity, and information security.
The most stable enterprise AI deployments keep systems tightly constrained. They rely on approved internal content, enforce role-based access, log usage, and require human sign-off. If IB Portal operates within these boundaries, it offers a realistic picture of how AI is reshaping investment banking.
Rather than replacing expertise, the BNP Paribas AI tool shows how AI can quietly enhance productivity by helping professionals navigate what their organisation already knows. For more updates on enterprise AI, financial services innovation, and real-world deployments, visit ainewstoday.org and stay ahead of the AI curve.
