Invest1 distinct publisher3 min readPublished
The Oslo firm co-founded by Kjell Inge Rokke has a 40 million euro first close and Halliburton, Investinor and KLP among its new backers. Fund I produced 24 investments and two exits.
The Investor · Invest desk

Compiled by The InvestorSomething wrong?How this is made
Forty million euros buys about 20 companies on the firm's own plan, spread over three to five years [1][10]. Initial cheques of 500,000 to 1 million euros put the cost of that programme at 10m to 20m euros, a quarter to half of the first close [11][3]. The remainder is reserve, and Fund I says the reserve gets spent: 24 initial investments generated 23 follow-ons, 47 decisions in total, close to one-for-one [12][1].
The pitch that assembled the new LP list is a multiple on other people's money. Fund I launched at 300 million kroner and its portfolio has since raised more than 2 billion kroner from outside investors, roughly 6.6 times the fund's original size [4][13][2]. That is a syndication record. It is a real skill, and it is not the same thing as money returned.
Tor Baekkelund's account of the sequence is unusually candid: a single backer was chosen so the team could act rather than spend years fundraising, with outside LPs always the intention [8]. Fund I's results then brought other industrial companies asking to join [9]. So the first fund bought the track record and the second fund sells it, which is the ordinary path a captive vehicle takes to become an institutional one.
The word "institutional" is doing some work here. Aker remains the principal investor, and Aker BP and Aker Solutions are on the LP list alongside Halliburton, KLP, Investinor and several Norwegian industrial families [6]. About 70% of investments to date sit in Norway and the Nordics, with the rest in Europe including WSense and Telgea [14]; hold that ratio and roughly 14 of the 20 Fund II companies land in one region [5]. A broader LP register does not by itself widen the aperture.
Sagar Chandna's chosen comparable is Germany's HV, an early backer of Neura Robotics [22], while the structural cousin in the source material is corporate venture, of the kind Emerald Technology Ventures runs with its 62 million dollar physical-AI fund with Japan's DIC [20]. The claimed difference is that RunwayVC's anchor is a family-controlled holding company rather than a strategic corporation [21]. That distinction is worth something only if the holding company's interest stays portfolio value rather than adjacency to its own operations, and Oyvind Eriksen's framing, citing Cognite and Nscale as proof that technology creates value when combined with deep industrial expertise, points at adjacency [16].
Fund II's opening cheques went to Minerva in humanoid robotics and HIVE in autonomy [7]. Both sit in categories where the round after the first one tends to be much larger than a five-person team writing million-euro cheques can lead [11]. Which means the thing under test in Fund II is not the thesis of connectivity, intelligence and autonomy that Chandna describes [17], but whether the syndication engine that carried Fund I's portfolio to 2 billion kroner still works when the buyers of the output are also the owners of the fund [19].
Ranked by verification strength, evidence, and original report placement.
RunwayVC's model includes limited partners such as Halliburton, Aker BP and Aker Solutions that are also potential customers of its portfolio companies.
RunwayVC has completed a first close of 40 million euros for its second fund, focusing on pre-seed and Series A companies in industrial AI, software, robotics, automation and autonomous systems.
The firm was founded in Oslo in 2021 by Tor Baekkelund and Kjell Inge Rokke, the billionaire chairman of Aker, and operates from the Aker Tech House in Fornebu.
The fund launched with 300 million Norwegian kroner, with Aker as the only backer.
The fund set up by Rokke, supported single-handedly by Aker since 2021, is for the first time accepting investment from parties other than itself.
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Single-source, firm-supplied figures with one unresolved discrepancy
All facts trace to one publisher's write-up of the firm's own fundraise, including named LPs, deal counts and portfolio capital. The numbers are specific and internally consistent, and the article does interview the partners directly, but nothing is independently verified, no LP commitment sizes or fund performance figures are given, and the source itself flags that only one of the two claimed exits is publicly listed.
Real deployed capital and named institutional LPs, early on outcomes
Adoption of the model is concrete rather than announced-only: a closed €40m tranche, a named roster of industrial and financial LPs, 24 initial plus 23 follow-on Fund I investments since 2022, more than NOK 2bn of external capital into portfolio companies, and two Fund II deals already done with one portfolio company (HIVE) reporting a $15m round and named industrial customers. What is missing is realised outcome evidence — only two exits, one publicly listed — so uptake of the vehicle is well ahead of proof of returns.
Institutionalisation framing runs ahead of verified performance
The story is framed as a family fund becoming institutional capital and as Fund I 'demonstrating the value of the model', but the supporting evidence is deployment volume rather than returns: two exits from 24 investments, one of them absent from the firm's own site, and no IRR or exit values. The €40m first close would fund only 25–50% of the stated 20-cheque plan at disclosed cheque sizes, and the LP-as-customer advantage is explicitly unresolved even in the source. Overstatement is moderate rather than severe because the headline capital, LP names and deal counts are specific and checkable.
Fundraise announcement amplified by a funding-news outlet
The material originates in a fund marketing moment: RunwayVC is raising toward a larger close, and the quotes come from its anchor investor's CEO and a new corporate LP's CEO, all of whom benefit from the model being seen as validated. The publisher is a startup-funding outlet whose beat depends on such announcements, and the piece carries interview access from the firm's partners. Mitigating factors are the outlet's inclusion of the exit discrepancy and its closing scepticism about customer-LPs.
Moderate-low: specific but uncorroborated single-publisher account
Confidence is limited by one source and one publisher covering a promotional event, with a self-acknowledged discrepancy on exits and no fund performance disclosure. It is not lower because the disclosed facts are unusually specific — named LPs, exact close size, cheque range, deal and follow-on counts — and because two Fund II investments and portfolio-level customer traction are independently checkable in principle.
build
The FCC closed the border on imported robots, so RoboStore is becoming a factory1 distinct publisher
build
Vivodyne is spending venture money on wet-lab throughput, not bigger models2 distinct publishers
invest
The humanoid supply chain is already Chinese: 18,500 of 19,100 units in six months1 distinct publisher
invest
Nvidia is brokering the Nordic build-out, not just supplying it2 distinct publishers
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 24, 2026