NBIM on AI cultural and organisational integration

Nicolai Tangen

By the evening of August 7, the same day GPT-5 was launched by Open AI, NBIM had it available to the entire organisation in a secure and scalable way. Joined on stage by CEO Nicolai Tangen at this year’s Arendalsuka, the team behind AI integration explains their aggressive approach.

Embracing the use of AI within Norges Bank Investment Management is clearly coming from the top, with CEO Nicolai Tangen, a former hedge fund manager, leading the charge.

“This is one of the most amazing things we are going to experience in our lifetime. It’s absolutely crazy,” said Tangen, speaking at this year’s Arendalsuka, Norway’s annual political gathering. “There are too many companies where it’s not happening enough, and if you’re not involved now, you’re falling behind and you’ll never get back on the offensive.”

Comparing himself to a “wasp” for “continuously irritating” staff on the matter – and likened to an AI “tornado” by NBIM’s chief technology and operating officer Birgitte Bryne who joined him on stage – Tangen said leading on AI integration involves attacking from all sides.

“Every time I get the microphone, every time we are doing something in the fund, I’m honking at everyone that ‘it’s AI that counts, it’s AI that counts, it’s AI that counts,’” he said.

There are tangible examples of AI’s deployment at the world’s largest investor, including an internally developed engine powered by AI to monitor and measure its portfolio managers’ skills, aiming to identify behavioural biases, improve decision making, efficiency of trades and save costs. (see How NBIM spots portfolio managers’ biases using AI)

Sponsored Content

Tangen has been on the record outlining the expected $400 million savings in trading costs per year, saying the fund has already achieved close to $100 million.

At Arendalsuka, he said the fund has now upped its goal to achieve a 20 per cent increase in efficiency in the hope that in two years NBIM will be almost “50 per cent more effective.”

And he’s vocal about the need for all employees to embrace AI – or get off the bus.

“We are not doing this to somehow save money, or fire people. On the contrary, we are doing this to achieve much more with the same resources. People [have to] understand that if they don’t do it, in the long run they will lose their jobs to the other person who uses AI because they are so much more efficient.”

In a presentation peppered with humour and infused with energy he said integrating AI gives employees purpose; it’s fun and is a chance to develop.

Pushing through organisational change

Taking the microphone, Bryne shared that the investor has introduced accessible training that includes mandatory 30-minute modules. This has helped create a level of knowledge amongst all employees whereby everyone “knows what we were talking about” and AI had stopped being “scary.”

NBIM has 40 specially trained AI ambassadors within the organisation. The investor recently ran a hackathon where all employees spanning offices in London, New York and Singapore gathered in small groups and were asked to solve specific tasks using AI, with the help of a technician.

Bryne said the investor has launched three tools, including a copilot/chatbot now used throughout the fund that allows staff to use AI without any knowledge of coding. Staff are supported though frequent “bumps” that often take three attempts to navigate. Meanwhile measuring AI gains includes monitoring which tools are most used, what tools people don’t use – and what it is they need to use.

“It’s very important feedback to have people themselves quantify what has helped them and what necessarily may not have helped them,” she said.

Fellow panellist Oscar Hjelde, who joined NBIM as a summer intern and is now a machine learning and AI engineer based out of NBIM’s London office, articulated the importance of “baking the cake internally” – aka in-house AI expertise. NBIM’s own tech stack enables it to be flexible introducing new solutions, and creates a proximity to the solutions the investor seeks, he said.

It has also helped build the investor’s developmental culture and allows AI integration to tap into NBIM’s “incredible number of talented colleagues.”

He said strategy involves taking the technologies “that we know are best, and make them available throughout the organization.” For example, GPT-5, the multimodal large language model developed and hosted by OpenAI was launched on August 7, 2025. That same evening “we had it available to the entire organization in a secure and scalable way.”

Hjelde reflected that the best way to predict the future is to build it – and the second-best way to predict the future is to talk to those who are building it.

It’s why the team regularly talk to Anthropic, the artificial intelligence startup backed by Amazon that allows them to “look into a crystal ball.” Proximity to companies at the cutting edge of AI also allows NBIM to have influence, and say to providers what is most important to them as a customer. “You get to be part of the latest adventure on the latest new technology and really sit at the forefront,” he said.

NBIM’s portfolio managers also regularly engage with portfolio companies on how they are investing in AI.

Bryne said that the next evolution will involve staff using technology that thinks for itself and carries out task “like an extra employee that is digital” in an “absolutely incredible boost for any business.”

It involves treating AI like a colleague and partner, rather than a tool.

Co-panellist Aleksander Stensby, founder of Norwegian AI company GritAI, said that when people think of AI as a tool they get “disappointed” when it doesn’t give the result they want. However, if they think of the technology as a partner they will give it feedback, and get the best results.

In this way people will move away from how they traditionally work on computers. Instead of typing on a keyboard and writing, they will live and talk to AI via microphones and cameras. These additional employees will share screens and press buttons in a multimodal future where they are assigned tasks.

“Nikolai can create his own little solution for whatever he struggles with during the day,” joked Stensby.

Integrating AI is not an IT project

Panellists reflected that some companies and their employees are progressive and forward-thinking AI natives. But many organisations have stalled because they lack a comprehensive strategy. They might have run pilot projects but have not strategically prioritised AI, and don’t have a belief or mission statement regarding the technology.

A successful strategy can also flounder because of a lack of urgency or shared ownership, which risks leaving people out in the cold.

Panellist Tine Austvoll Jensen, country director of Google Norway and board member, said that geopolitical unrest has also injected caution into AI strategies amongst Nordic companies concerned about the safety of American cloud solutions. Although Google and its competitors have solutions on European soil, she said uncertainty has held companies back.

Panellists concluded that such is the pace of development last year was “like the Stone Age” in relation to where AI is today.

Leave a Comment

Pension funds confront the question of who owns AI

Pension funds confront the question of who owns AI

As the use of AI within asset owners evolves, organisations are grappling with the governance question of where the strategy and accountability sit. Darcy Song looks at the treatment of AI organisationally within a number of high-profile funds, including OTPP, AustralianSuper, CPP and Norges Bank.

Sort content by

CPP Investments on how AI redefines core investing roles and processes

CPP Investments’ trials show AI agents can handle key investment tasks end-to-end. In an interview, chief operating officer Jon Webster says tight governance, and the right human oversight, is the difference between breakthroughs and mistakes.

HESTA prepares AI investment framework for total fund clarity

The A$98 billion ($64 billion) Australian super fund HESTA is laying the groundwork for a more systematic framework for using AI across its total portfolio, solidifying use cases in research, forecasting, risk management and private assets that all centres on the objective of “seeing risks earlier and clearer”.

Responsible AI: Railpen lays out the risks

Much is written extolling the investor opportunities inherent in AI at a time policy makers continue to prioritise deregulation and innovation over safety, but a new report from £34 billion Railpen on the risks AI holds for investors' portfolio companies provides a valuable reality check.

AP4: Why a dynamic, shorter term allocation is paying off

Volatile markets have provided a rich hunting ground and opportunistic best ideas have come thick and fast for AP4’s new five-pronged global allocation made up of systematic equity, currency and rates, asset allocation, hedge funds/external mandates and analysis. Magdalena Högberg explains the risks and opportunities of the best ideas allocation.

Large language models to spark ‘sea change’ in investment analysis

Andrew Lo, finance professor at the MIT Sloan School of Management, believes large language models can bridge the gap between fundamental and quantitative investing in a way that was unfathomable five or 10 years ago, and create ‘quantamental’ investment strategies which would bring together the best of both worlds.

Investors spot key AI applications in portfolio management

AI can add value in almost every part of the investment process, including manager selection and monitoring and risk management, providing support in information gathering and analysis, sometimes from non-traditional and hard-to-access data sources, according to a new CFA Institute Research and Policy Center report.