SuperSeed Capital Ld - Q4 2025 Results

About SuperSeed Capital Limited

SuperSeed exists to back Europe's best B2B SaaS founders at the earliest stages and to help them build great companies. In the short term, our portfolio companies enable their customers to drive revenue growth and efficiency savings using next-generation software and AI. In the long-term, they have an opportunity to create category defining global technology companies. SuperSeed focuses on the fundamentals by helping founders build good companies with strong unit economics and sensible distribution models.

Forward-looking statements

This announcement contains statements that are or may be forward-looking statements. All statements other than statements of historical facts included in this announcement may be forward-looking statements, including statements that relate to the Company's future prospects, developments and strategies. The Company does not accept any responsibility for the accuracy or completeness of any information reported by the press or other media, nor the fairness or appropriateness of any forecasts, views or opinions expressed by the press or other media regarding the Group. The Company makes no representation as to the appropriateness, accuracy, completeness or reliability of any such information or publication.

Forward-looking statements are identified by their use of terms and phrases such as "believe", "targets", "expects", "aim", "anticipate", "projects", "would", "could", "envisage", "estimate", "intend", "may", "plan", "will" or the negative of those, variations or comparable expressions, including references to assumptions. The forward-looking statements in this announcement are based on current expectations and are subject to known and unknown risks and uncertainties that could cause actual results, performance and achievements to differ materially from any results, performance or achievements expressed or implied by such forward-looking statements. Factors that may cause actual results to differ materially from those expressed or implied by such forward looking statements include, but are not limited to, those described in the Risk Management Framework section of the Company's most recent Annual Report. These forward-looking statements are based on numerous assumptions regarding the present and future business strategies of the Group and the environment in which it is and will operate in the future. All subsequent oral or written forward-looking statements attributed to the Company or any persons acting on its behalf are expressly qualified in their entirety by the cautionary statement above. Each forward-looking statement speaks only as at the date of this announcement. Except as required by law, regulatory requirement, the Listing Rules and the Disclosure Guidance and Transparency Rules, neither the Company nor any other party intends to update or revise these forward-looking statements, whether as a result of new information, future events or otherwise.

Investment Manager's Review

The Capability Constraint is Over

The technical community has been using Claude Code for a year. It's an agent harness that enables Claude to execute your instructions: not just answering questions, but taking actions on your computer, connecting to data sources, accomplishing objectives. Originally built for writing code, it now works for marketing, email processing, research, and everything else.

Three things happened in January that changed the game:

First, Anthropic released CoWork. This puts Claude Code-like capabilities in the hands of everyone, not just developers. There's been a sudden realisation of just how powerful the agentic way of working is, and now anyone can access it.

Second, the OpenClaw project takes the approach further. Claude Code is still "give me a prompt, go do actions, come back with results." OpenClaw is "here are my objectives for the world, go figure out how to achieve them, even if it takes hours, days, or weeks." You could ask it to produce a podcast: research guests, contact them, create background research, book them in your calendar, reply to their emails, reach out on WhatsApp. Anything you can accomplish through a computer; the agent can try to do. That doesn't mean it will succeed. And many people have trepidation about giving an agent full access to their email, WhatsApp, passwords, everything. But this is the agentic future, and we've just seen a powerful glimpse of it.

Third, China's Moonshot released Kimi K2.5, an open source model that is comparable to Anthropic's Opus 4.6 at agentic work. It's built for parallelisation: spinning up a hundred agents to run simultaneously, all orchestrated by one model. Match that with the OpenClaw harness and you can see how powerful this becomes.

The capability constraint is over. The models are here. The harnesses work. And yet 95% of enterprise AI projects still fail to reach production.

The Real Bottleneck

The failure isn't technical. MIT's 2025 research found the top obstacles are data quality (43%), lack of technical maturity (43%), and skills shortage (35%). S&P Global reports 42% of companies now abandon AI initiatives before production, up from 17% the previous year.

Enterprises have access to powerful models but lack the infrastructure to deploy them. The constraint has shifted from "can we build AI that does this?" to "can we tell if AI did it correctly?".

Software development transformed first, but not because we have systems to check AI output today. It's because the models were trained on those systems. During training, models learned from test suites via reinforcement learning. They got fast feedback on what works and what doesn't. Software has right and wrong answers, unlike so many other fields. The models were beaten into perfection by having first-class, fast feedback during training.

Other sectors lack this training infrastructure. Institutional knowledge lives in people's heads and dies when they leave. Decision logic sits in spreadsheets no one maintains. There's no equivalent of a test suite to train models against, because those verification systems were never built.

Where Value Concentrates

We've written before about the broken economics of AI infrastructure. A $30/month Cursor subscription flows entirely to Anthropic, who loses money on every transaction paying hyperscalers, who lose money funding the frontier labs who buy their chips. Only NVIDIA profits. The funding loop masks dysfunction at every layer.

As these economics resolve (inference costs collapsing, open-source matching proprietary at one-tenth the price), value will migrate up the stack. The PC era shifted value from hardware to software. The internet era shifted value from infrastructure to applications. The AI era will shift value from model providers to those who make AI deployable.

The winning companies won't build bigger models. They'll build applications that unlock sector-specific value, with the decision scaffolding and verification that makes AI outputs trustworthy.

This is where the Fund's portfolio sits.

Q4 2025 Fund Portfolio Performance

While the industry debates whether the AI bubble will burst, the Fund's portfolio companies are building applications that transform specific industries. They encode domain expertise, verification systems, and decision infrastructure into software that makes AI actually work.

Fund Metrics

Q4 was the strongest quarter to date, capping 2025 as a strong year:

·     Net IRR: 21.1%

·     TVPI: 1.25x

·     DPI: 0.13x

·     Portfolio CARR: $22 million (up 69% from $13 million at start of 2025)

Fund portfolio revenue grew at nearly 100% annualised in Q4.

Over four years, the Fund has delivered 21.1% net IRR compared to 10.6% annualised from holding the S&P 500 (49.4% total return for the S&P 500 over the period). As the Fund's portfolio matures and companies progress through subsequent funding rounds, we expect continued outperformance.

Investment Period Complete

After four years, the Fund's investment period is complete. The Fund has built a portfolio of 27 companies, each applying AI to transform specific sectors.

The timing matters. The Fund began deploying in January 2022, nearly a year before the release of ChatGPT. It invested through the generative AI revolution, watching capabilities explode while deployment remained stubbornly difficult. Now, as the Fund closes its investment period, the industry is finally grappling with the gap between what AI can do and what organisations can verify.

New Investments

Q4 saw two new investments for the Fund, with two more closing in early Q1 2026:

Hive Autonomy is a universal platform for remote and autonomous operation of any industrial vehicle. One operator can control multiple machines simultaneously from a centralised station. Removes workers from dangerous environments, improves productivity. Forklifts, excavators, diggers. These machines move trillions of pounds of materials globally and operate exactly as they did fifty years ago.

SOLVE Chemistry combines automated high-throughput flow chemistry with machine learning. They compress pharmaceutical process development from 9-24 months to weeks, generating 20x more data than traditional methods. Already working with BASF, AstraZeneca, and Hovione.

In early Q1 2026, the Fund completed investments in Cursive (a DeepMind spin-out working on real-time generative AI models, still in stealth) and All3 (construction technology extending our Physical AI thesis into one of the world's largest and least digitised industries).

Portfolio Spotlight: Physical AI

Three of the Fund's investment portfolio companies show where AI creates defensible value in the physical world:

Hive solves implementation where traditional automation fails. Unlike AGVs requiring controlled environments and fixed paths, Hive's AI operates in unpredictable conditions: active warehouses, construction sites, industrial facilities. Every site differs. The expertise isn't in the model. It's making autonomy work where scripted automation breaks down.

OctaiPipe now has ABB as both investor and distribution partner. ABB offers a strong distribution channel for Octaipipe.

All3 applies similar logic to construction. Paper processes, tribal knowledge, and manual coordination waste billions annually. The decision infrastructure that makes AI deployable in construction looks nothing like what works in software. All3 is building it.

Looking Forward

Several of the Fund's investment portfolio companies are progressing toward Series A. We expect significant activity in H1 2026.

The AI revolution accelerates. Models grow more powerful weekly. Open source narrows the gap with proprietary systems. Agentic architectures mature from research to production.

But capability was never the constraint. The companies that win will build applications that make AI actually deployable in specific sectors, with the domain expertise and verification that transforms powerful models into trustworthy software.

That's what we built the Fund's portfolio to do.

Related Party Transactions

Convertible Loan Note Instrument

Further to the announcement made on 26 June 2024, the Company is pleased to announce it has extended the duration of the loan facility, for the issue of convertible loan notes (the "Convertible Loan Notes") to the Investment Manager. As and when issued in the future, the proceeds of the Convertible Loan Notes shall be used to fund the Company's short-term working capital and capital expenditure requirements.

The Convertible Loan Note instrument has been amended such that: 1) the aggregate principal amount of the Notes outstanding at any time has been reduced from £1,000,000 to £400,000; and 2) the latest date on which any Convertible Loan Notes in issue, together with accrued interest, will be redeemable has been extended from 21 June 2026 to 12 February 2028 (the "New Redemption Date").

The Convertible Loan Notes shall automatically convert, in part or in full, at any time from the date of issue until the New Redemption Date, into fully paid ordinary shares in the Company at a price of £1.30 per Share on written notice of the Noteholder. Interest shall be payable on any outstanding Convertible Loan Notes at a rate expected to be 10% - 15% per annum (subject to changing market rates) calculated based on a 365-day year and the actual number of days elapsed from the date of issue of the Convertible Loan Notes to the New Redemption Date.

The Investment Manager is a related party of the Company pursuant to the rules of the Aquis Stock Exchange. The independent directors of the Company, having exercised reasonable care, skill and diligence believe that the extension of the Convertible Loan Notes is fair and reasonable as far as the shareholders of the Company are concerned.