Moderator's Introduction
Moderator:
2026 is not going to reward the "trendy industry," but rather those who build businesses that turn trends into revenue. The research base is clear: AI, biotechnology/digital health, renewables/CleanTech, semiconductors, and cybersecurity stand out as winners due to estimated high margins (AI up to +30%), record investment (AI captured 64% of global VC in 1H 2025), and regulatory pressure (Green Deal, privacy, digital resilience). However, the real debate is not the list but the "how": what business models withstand competition, regulation, and commoditization. There’s also an uncomfortable tension: sectors like oil or banking may perform due to geopolitical shocks, even if they aren't the dominant narrative. Today we have a triologue: Francisco Torres (profitability and efficiency without theatrics), Tomás Rivera (validation and products people pay for), and Sofía Valenzuela (model architecture, fit, and cash flow). We will discuss winning industries, but most importantly, opportunities that an entrepreneur should seize in 2026.
---
Opening Round
Francisco Torres:
If you ask me about "winning industries," AI and cybersecurity top the list for an operational reason: they allow the sale of measurable efficiency in weeks, not five-year promises. AI absorbing 64% of global VC in 1H 2025 does not excite me because of hype; it interests me due to the cost structure change: smaller teams delivering more. The typical mistake is to confuse "AI" with "making a wrapper." In 2026, those who package AI in processes where clients already pay will win: support, sales, compliance, financial back office, logistics. Success also belongs to those not reliant on subsidies. In digital health and biotech, I see margins, yes, but the sales cycles and regulation: the winning model there is B2B with hospitals, insurers, or labs, clear contracts and evidence. In CleanTech, the business isn't just the panel; it's financing, installation, maintenance, and energy management with recurring contracts. Without cash flow and a healthy economic unit, there is no "winning industry."
Tomás Rivera:
I don’t buy the “winning industry” idea as if it were a stock. In 2026, the teams that quickly validate real willingness to pay and adapt will win. That said: AI and cybersecurity have a massive tactical advantage for experimentation because the customer feels pain today. With AI, the repeated pattern is: they promise productivity, but no one pays when integration is a nightmare or the result is not auditable. So the opportunities that cannot be wasted are hyper-specific: vertical copilots with hard metrics (time saved, resolved tickets, churn reduction), and pricing that captures value from day one. In cybersecurity, the "AI-boosted threat" makes many SMEs buy out of fear, but they leave if there’s no onboarding and response. In CleanTech, demand grows due to regulation, yes, but the product bought is "savings" and "compliance", not abstract sustainability. In digital health, if there’s no paid pilot with a hospital or insurer, it’s just PowerPoint. 2026 is defined by paid experiments, not presentations.
Sofía Valenzuela:
I view this as blueprints: in 2026, models with pieces that don’t fit collapse, even if they're in a sexy industry. AI, biotech, CleanTech, semiconductors, and cybersecurity have tailwinds, but wind doesn’t replace structure. The question is which business architectures can hold weight: specific segments, efficient channels, upfront payment or recurrence, and controlled variable costs. AI has an estimated margin of +30%, but also the risk of commoditization: if your pitch is "I use a model," your building won't have columns. The column is the data, the workflow, regulatory compliance, and integration with existing systems. In CleanTech, the critical piece is financing and measurement: performance-based contracts, PPAs, maintenance, and monitoring. In digital health, fit depends on regulation, clinical evidence, and who pays: hospital, insurer, or patient. In cybersecurity, the winning architecture combines prevention + response + compliance, packaged in a subscription. Companies don’t win by industry; they win by fit and cash flow.
---
Debate Round
Moderator:
I’ll start with friction: AI concentrates investment and margins, but also saturation of indistinguishable "copilots". Francisco, you advocate for efficiency and sales from day one. Tomás, you say "paid experiment." Sofía, you discuss structural columns. What specific opportunity in AI 2026 doesn’t become a commodity in 6 months?
Francisco Torres:
The opportunity isn’t "chat with AI", it's the redesign of the operation with a KPI that the CFO recognizes. Example: automation of accounts receivable and reconciliation for SMEs and mid-market, with integration to ERPs and banks. Here, the product is reduced DSO, fewer errors, and less incremental headcount. This sells with a monthly contract and closed onboarding, not with endless freemium. Another: compliance and internal auditing assisted by AI in regulated sectors. It’s less glamorous but less substitutable. The one who wins is the one who controls computing costs, has a playbook for implementation, and charges for results or by seat. If you depend on "being smarter with prompts," you’ll get copied. If you sell process + integration + support, you last.
Tomás Rivera:
I agree with "process", but if you don’t validate it as immediate pain, you’ll be stuck spending a year integrating. I’ve seen teams sell "AI for finance" and then discover the accounting team can't change their flow due to internal compliance. Non-commodity opportunity: tools that plug into a specific action and charge per transaction or event. Example: automatic classification of incidents and draft responses in customer service with measurable SLAs, or fraud detection in e-commerce with direct recovery of losses. If the customer sees the delta in a week, they pay and stay. If you ask for a complete re-architecture of the ERP, it kills the sales cycle. The advantage in 2026 is the speed of learning with paid pilots and controlled outreach.
Sofía Valenzuela:
Both of you touch on real pieces, but one is missing: channel and surgical segmentation. "Accounts receivable" for "SMEs" is too broad; it’s like designing a bridge for any river. The non-commodity is born when you atomize: for example, accounts receivable for private clinics billing insurers, or for logistics with a high dispute rate. There, the data and rules are specific, integration becomes repeatable and CAC drops with specialized channels. In AI, the column is repetition: same segment, same workflow, same playbook. If each implementation is bespoke, the +30% margin evaporates in services. The winner in 2026 is whoever converts implementation into a product and sales into a system.
---
Moderator:
Second friction: CleanTech and renewables have regulatory tailwinds but estimated net margins lower (15%) and high competition. Is it still a “winning industry” for startups or a territory for incumbents?
Francisco Torres:
As an industry, yes. As a startup, it depends on the model. Heavy hardware and CAPEX without self-financing is a trap. However, energy services with recurring contracts and clear billing is a business. In Spain, for example, solar installation is competitive; what’s profitable is the complete package: auditing, processing, installation, maintenance, and monitoring with a fee. If you can additionally add demand management and batteries, the ticket increases. CleanTech for me is an "execution business": less pitch, more operations, more agreements with installers, more after-sales. And it's financed by clients, not by rounds waiting for capital costs to drop.
Tomás Rivera:
The problem is many enter CleanTech with narrative and no proof of willingness-to-pay. The real opportunity is to sell verifiable savings. If you promise a 20% reduction in energy, show it on a bill in 30 days with a paid pilot. Efficiency products for businesses and SMEs: sensors, control, HVAC optimization, predictive maintenance. And beware: "ESG consulting" can be smoke if it doesn’t tie delivery to compliance or tenders. If the client’s incentive is to pass audits or access funding, there is a price. If it’s just “feeling green,” it’s fragile. In 2026, winners are those who package ROI in a simple, replicable contract.
Sofía Valenzuela:
CleanTech is a winner when the model is designed for execution margins, not lab margins. The building stands if cash flow comes in before costs crush you: upfront billing, associated financing, or performance contracts where you measure and share savings. If you depend on changing subsidies, your structure has a moving beam. And for startups, the advantage is not to compete with panel manufacturers but to occupy the system's "gaps": demand aggregation, maintenance, monitoring software, asset management, and compliance. Here the channel can be installers, property managers, or banks financing energy reforms. The model wins when those pieces fit together seamlessly.
---
Closing Round
Francisco Torres:
In 2026, applied AI, cybersecurity, and energy efficiency win when sold as operational results, not technology. The missed opportunity is building "process products": integration, playbook, support, and recurring pricing from the first client. Biotech and digital health can yield margins, but with B2B contracts and evidence, not generic apps. The most reliable signal is simple: a paying client, controlled variable costs, and an implementation that becomes repeatable. The rest is noise. The real competitive advantage is profitable execution.
Tomás Rivera:
Industries won’t save you; learning faster than the market without burning cash will. In 2026, AI and cybersecurity facilitate high-impact paid experiments, and CleanTech allows you to sell savings and compliance if measured in weeks. The impossible-to-miss opportunity is to design visible-price offerings from day one, short pilots, hard metrics, and natural expansion. Anything that requires faith, infinite integration, or three-year plans collides with reality. Real growth only occurs when the illusion of a perfect plan is abandoned, embracing constant validation with real clients.
Sofía Valenzuela:
I see 2026 as an engineering test: models with snug-fitting pieces win, not slogans. AI wins when data, workflow, and channel are specific and repeatable. CleanTech wins when billing is tied to measurement and cash flow is designed to support execution. Digital health wins when it’s clear who pays and why, with integrated evidence and compliance. The missed opportunity is atomization: one segment, one channel, one proposition, a cash machine. Companies don’t fail for lack of ideas, but because their model pieces cannot fit together to generate measurable value and sustainable cash flow.
