Artificial intelligence is changing how venture capital firms discover startups, evaluate opportunities and support portfolio companies. But as AI takes on more of the analytical and operational work, what remains uniquely human in investment decisions?
In this exclusive interview with Global Apex Tech, Noor Sweid, Founder and Managing Partner of Global Ventures, shares how her firm is integrating AI into its investment operations, from startup sourcing and portfolio monitoring to founder support.
She also discusses the importance of human judgment, the risks of AI-driven analysis and what these changes could mean for founders seeking investment across emerging markets. Drawing on practical examples from Global Ventures, the conversation explores where AI is delivering measurable value today and how the relationship between investors and founders may evolve in the years ahead.
1. How has AI changed the way your firm finds, evaluates or supports startups over the past 12–18 months? Which change has been most consequential in practice?
AI is now part of how we work every day at Global Ventures, and the change begins with sourcing. Adding companies to our CRM used to mean entering every field by hand, one company at a time. Today we can share a demo day list, ask which companies we already track, and have the new ones created with owner, status, round, market and source completed, ready for screening calls.
After we invest, AI strengthens the value creation work we have been building since 2018 through our PARTNER framework, which spans processes, awareness, revenue growth and expansion, talent and recruitment, network, exit pathing and raising. Portfolio updates that once arrived by email, WhatsApp, calls and spreadsheets now come together in one live view, and our internal AI, Blue Dot, prepares board meetings and drafts the one-pager.
The most consequential change has been in our network. Around 15,000 relationships once sat in separate address books. They can now be searched together, which supports our sourcing and helps a founder who needs a customer. Blue Dot suggests the connection, and our team applies judgment and makes the introduction. What once relied on chance is now a repeatable process.
2. Where in the investment workflow—sourcing, screening, diligence, investment committees, portfolio monitoring or LP reporting—is AI creating genuine value today? What evidence tells you it is improving the work rather than simply making it faster?
The clearest value today is in sourcing, portfolio monitoring and portfolio support. A useful way to look at it is through three shifts: capacity, cadence and initiative, a framing from a 2026 Stanford Graduate School of Business note on AI in emerging-market venture capital that draws on our firm among others.
In sourcing and screening, the gain is capacity. Getting companies into our pipeline takes a fraction of the time, and we can check a full list of start-ups against our own records in seconds, so we know which founders we already have history with before the first call. In diligence and investment decisions, the industry is still learning where AI adds the most.
After we invest, the gain is cadence. We once learned how a company was doing at a board meeting or in a monthly update. With portfolio data in one live view, and a named owner for every value creation project, we see change between meetings. That is the evidence that the work is better as well as faster: we know sooner, and the right person acts. Initiative, where a system flags an issue before a founder raises it, is at an earlier stage and depends on clean, consistent data.
3. Which parts of venture investing should remain decisively human, even as AI capabilities improve? How do founder relationships, market context and judgment affect that boundary?
The decision to back a founder should remain human. AI is very good at showing what a company has done so far. How a founder will respond when the plan changes is something we learn from time spent together. Grit and resilience do not show up in a data room, and they often decide the outcome.
Market context is part of that decision. We invest in emerging markets, where data can be limited or out of date and may not show how a regulator approaches a sector or how customers actually pay. This is where our judgment needs to be strongest.
After we invest, a similar balance applies. For context, a 2025 London Business School case study on our firm recorded our estimate that about 80% of the value we add comes from systemized, repeatable support, while the remaining 20%, bespoke and in real time, tends to drive the outcomes that matter most. AI is taking on more of the 80%, which frees our team for the 20%. A system can suggest who a founder should meet. A person decides whether to make the introduction, because it spends trust that took years to build.
"AI now carries much of the work at the top of our funnel. Deciding which founders to back, and supporting them after we invest, is still a very human part of venture." — Noor Sweid, Founder and Managing Partner, Global Ventures
4. How do you manage the risks of incomplete data, bias, hallucination, confidentiality and increasingly similar machine-generated analysis? Who is accountable for reviewing or challenging an AI-assisted conclusion?
Our starting point is simple: we use AI for what it does best and keep human quality control. Systems prepare and recommend, and people remain responsible for decisions that affect founders, employees, capital or reputation.
In sourcing and screening, incomplete data is the main risk in our markets. A system is only as good as the record behind it, so we work to keep ours clean and current. On hallucination and bias, figures and claims that matter are checked against their source by a person. Sameness is a further risk: if every firm asks the same models the same questions, every firm receives the same memo. What differentiates an investor is what it learns from founders and from its network.
After we invest, confidentiality begins with consent. We want founders to know what information we collect and analyze, and how one company's information may be used elsewhere in the firm. Anything drafted with AI is reviewed by someone on our team before it reaches a founder.
Accountability stays with people. The team member who brings a recommendation to the partnership is responsible for it, whichever tools helped prepare it, and the partners are responsible for the decision.
5. Please share a non-confidential example in which AI materially influenced a sourcing, diligence or portfolio-support decision. What changed, what remained a human call, and what did the outcome teach you?
A recent example comes from sourcing. We see around 1,500 opportunities a quarter and back only a few, so the way companies enter our system matters. After a recent demo day we had a list of 37 start-ups. Previously, someone on the team would have entered each one into our CRM by hand, field by field, before we could begin thinking about who to call.
This time we shared the list with our AI and asked what we already had. Within seconds it showed that 11 of the 37 were already tracked and 26 were new. With one more prompt it created the 26 with owner, status, round, market and source completed, ready for screening calls.
What changed was how the team spent its time. More of it went to the screening calls themselves, and we could approach the 11 companies we already knew with our history in front of us. By our own estimate, this kind of work has gone from about five hours a week to around 30 minutes.
What remained a human call was everything that followed: which founders to call first, what to ask them, and whether any of them fit what we look for. The AI did not assess quality. That stayed with the team.
The lesson is that much of the value comes from memory and coverage. A firm that knows what it already knows makes better use of every conversation, and a system holds that memory more reliably than a busy team. It also reinforced how much the quality of our records matters once AI relies on them.
6. Across APAC’s fragmented markets, how do you expect AI to change access to capital, investor differentiation and founder–investor relationships over the next 12–24 months? What should founders understand now?
AI is reaching every market, which is what makes this moment so promising. The pace varies by market, and adoption is picking up quickly across all of them, because the same tools are available to founders and investors everywhere. Fragmentation shapes how the change shows up. The direction is the same. We have always believed that exceptional founders can be found anywhere, and AI makes that belief easier to act on.
On access to capital, we expect AI to widen the top of the funnel. Founders outside the main hubs can be discovered and screened by investors who might never have met them, and some can now reach the same milestones with less capital. As research and market mapping become easier, investors will start to look more alike in their analysis.
Differentiation then moves to what happens after the investment. AI can help a small team offer its founders more research and monitoring, while relationships, introductions and support through a difficult period still depend on people.
Looking ahead, founders should keep their company's public record accurate, because more early screening is done by systems, and ask other founders what an investor did for them after investing.
Looking Ahead: AI and the Future of Venture Capital
As AI becomes more integrated into venture capital, its value may extend beyond efficiency to helping investors identify opportunities earlier, better understand their portfolios and provide more consistent support to founders.
Noor Sweid’s experience at Global Ventures highlights an important distinction: while technology can improve how investment firms operate, human judgment, trust and relationships remain central to investment decisions.
For startup founders, the message is equally relevant. As AI makes companies more discoverable to investors, maintaining accurate business information, building strong relationships and choosing investors who offer meaningful support beyond capital will remain important.
About Noor Sweid
Noor Sweid is the Founder and Managing Partner of Global Ventures, a Dubai-based, Series A-focused venture capital firm with more than $400 million in assets under management and over 70 portfolio companies. Before founding the firm in 2018, she scaled Depa tenfold and took it public in 2008, and founded ZenYoga, which was acquired in 2014. She is the Founding Chairwoman of the Middle East Venture Capital Association and the author of Coming of Age. Noor holds an MBA from MIT Sloan and degrees in Finance and Economics from Boston College.
Noor Sweid · Founder and Managing Partner, Global Ventures · Dubai, United Arab Emirates · Global Ventures website · LinkedIn profile
Published by the Global Apex Tech Editorial Desk. Partner involvement, when applicable, is disclosed above the headline. For editorial questions or source material, contact editor@globalapextech.org.