Vertical AI Market Map 2026: Israeli Startups and Funding Rounds

Views: 98

For the past three years, investors have predicted that the future of AI would be vertical. In the first half of 2026, that thesis started to become visible in Israel’s funding data.

Israeli startups raised billions of dollars during the period, but the more interesting story isn’t the topline number, it’s where the application-layer capital went. Banks, insurers, defence organisations, property managers, pharmaceutical companies and gaming studios are no longer buying AI to summarise information. They’re beginning to deploy systems that underwrite policies, manage loans, discover drugs, operate buildings, direct traffic and execute entire workflows end to end.

That’s the shift from vertical SaaS to vertical AI. Traditional software helped professionals organise their work. This generation is being paid to perform it.

But the map also requires some discipline. Not every startup using an LLM is a vertical-AI company, and not every agent platform owns a vertical. This piece tracks Israeli companies that announced funding in H1 2026 (January–June) and separates industry-specific applications from horizontal AI infrastructure, functional copilots and cybersecuritym categories that get lumped into “vertical AI” maps far too often.

What actually counts as vertical AI

A company should qualify as vertical AI when it meets four criteria:

1. It serves a recognisable industry or profession.
2. Its product encodes domain-specific workflows, data or regulation.
3. It performs or substantially automates work, rather than adding an AI interface on top of existing tasks.
4. Its competitive advantage would weaken meaningfully if the domain expertise were removed.

That framework splits the AI landscape into three layers, and only one of them is the core map:

Cybersecurity gets its own panel, not a slot on the map. It’s one of Israel’s largest technology sectors, but most AI-native security products sell horizontally across every industry rather than encoding one profession’s workflow. Folding all of Israeli cyber into “vertical AI” would overwhelm the map and make the category meaningless. Think of AI-native security as a large, important application category that sits adjacent to vertical AI, not inside it.

The H1 2026 map, by industry

Every company below was checked against its own funding announcement: press release, TechCrunch, Calcalist/CTech or Globes coverage, rather than taken at face value from secondary trackers. Several rounds turned out to be reported inconsistently across outlets or headlined with cumulative totals rather than the actual new round size; corrections are noted inline.

Finance, fintech and insurance – the broadest category

Nine companies, $196M disclosed, rounds spanning $6M to $70M with a roughly even spread across seed, Series A and later stages. No single round captured more than a fifth of the category’s capital, a sign of genuine breadth rather than one hot deal.

• Lama AI – $12M Series A (June 2026, led by EJF Ventures) – AI agents that underwrite lending for community and regional banks. Total funding now exceeds $20M; some outlets miscited the round as $10M or $20M, but $12M is the corroborated figure.
• Honeycomb Insurance – $40M growth round (June 2026, led by Zeev Ventures) – AI-native underwriting specifically for apartment buildings, condo associations and commercial multifamily property, not insurance broadly. Total funding to date: $95M.
• Novella – $16M Series A (May 2026, led by Brewer Lane Ventures) – AI brokers for excess-and-surplus (E&S) insurance. The often-cited $21M figure is cumulative funding, not the new round.
• Hypercore – $13.5M Series A (February 2026, led by Insight Partners) – an “AI Admin Agent” that runs loan operations for private-credit funds.
• Arito – $6M Seed (2026, led by Amplify Partners) – agentic analytics and BI for finance and revenue teams.
• Datarails – $70M Series C (January 2026, led by One Peak) – a FinanceOS platform for FP&A and financial analysis. Total funding: $175M.
• Atlas Invest – $25M in combined equity and debt (January 2026, led by BlackRock) – commercial real-estate bridge lending.
• Viewz – $7M Seed (May 2026, led by Ibex Investors and FLINT Capital) – ledger automation and a “continuous close” finance-ops platform.
• Dono – $6.5M Seed (February 2026, led by Link Ventures) – property-record and title infrastructure. Total funding: $10.2M.

Healthcare and life sciences – active, but still the smallest core category

Four companies and $59.4M disclosed — roughly 8% of all core vertical-AI capital tracked here. That’s a notable contrast with the US market, where Menlo Ventures estimates healthcare accounted for close to half of all vertical-AI spending in 2025. Healthcare also splits cleanly into two sub-categories that behave differently: clinical AI, which still runs into regulatory and patient-safety friction, and administrative AI, which is commercialising faster because it doesn’t.

Clinical:

• Scala Biodesign – $16M Series A (March 2026, led by Grove Ventures) – an AI protein-design platform (ScalaOS), spun out of Weizmann Institute research.
• Converge Bio – $25M Series A (January 2026, led by Bessemer Venture Partners) – GenAI drug discovery, founded by former Meta, OpenAI and Wiz executives.

Administrative:

• Take2 – $14M Series A (February 2026, led by Human Capital) – autonomous AI agents that handle healthcare recruiting and interviewing.
• CareBestie – $4.4M Seed (2026, led by TLV Partners) – voice-based monitoring and engagement for home-care and hospice patients.

Defence, security and public infrastructure: where the biggest rounds live

Six companies, $361.7M disclosed — over half of all core vertical-AI capital in this map, even though it’s the smallest company count. This is Israel’s clearest domain advantage showing up in the funding data: battlefield intelligence, autonomy, electronic warfare, and physical infrastructure like traffic and weather systems.

• Airis Labs – $31M Series B (May 2026, led by PSG Equity), on top of an earlier seed and $11M Series A, bringing total disclosed funding to $60M – AI-powered analysis of battlefield video and visual intelligence.
• Shifters – $10.2M Seed (June 2026, led by Ace Capital Partners) – autonomous ground robots for defence.
• Tenna Systems – $13.5M Seed (February 2026, led by Costanoa Ventures) – sensor fusion for electronic warfare. (Branded “Tenna Systems” to distinguish it from an unrelated US fleet-tracking company also called Tenna.)
• Conntour – $7M Seed (March 2026, led by General Catalyst and Y Combinator) – natural-language search across video surveillance systems.
• NoTraffic – $90M Series C (March 2026, led by PSG Equity) – autonomous, adaptive traffic-signal management. Total funding: $165M.
• Tomorrow.io – $175M Series F (February 2026) plus a $35M follow-on (May 2026) – weather intelligence and a satellite constellation. Combined: $210M, making it the single largest round in this map.

Several of these blend vertical software with robotics and physical hardware — worth calling out rather than forcing everything into a SaaS taxonomy.

Commerce, advertising and media

Four companies, $77.2M disclosed.

• Sett – $30M Series B (April 2026, led by Greenfield Partners) – AI agents for game marketing and user acquisition.
• Rep AI – $6.2M follow-on (May 2026, led by Silicon Road Ventures, with Zendesk’s first-ever Israeli startup investment) – ecommerce customer engagement agents.
• Brandlight – $30M Series A (February 2026, led by Pelion Venture Partners) – monitoring brand visibility across AI-answer engines like ChatGPT and Perplexity.
• Recapp – $11M (February 2026, led by Liquid2 Ventures) – personalised, AI-generated sports highlights.

Real estate and the built world

This category is already credible on its own, and it overlaps with finance and infrastructure above: Atlas Invest (real-estate lending), Dono (property records) and NoTraffic (urban mobility) all belong here too. The clearest dedicated example:

• Visitt – $22M Series B (January 2026, led by Susquehanna Growth Equity) – AI for commercial-property operations, reporting 900% growth in managed square footage in 2025.

It’s a category worth watching alongside Bessemer’s built-world AI roadmap, which flags construction and real estate as underinvested relative to their share of GDP.

Functional AI: important, but not a vertical

These companies sell into a job function rather than an industry, and conflating them with vertical AI is exactly the kind of category error this piece is trying to avoid. Six raised in H1 2026 for a combined $194M — comparable to the entire commerce and healthcare categories combined, and more concentrated at the top: the three largest functional-AI rounds captured 70% of the category’s capital, compared with roughly half in core vertical AI.

• Winn.AI – $18M Series A (February 2026, led by Insight Partners) – real-time AI sales guidance.
• Notch – $30M Series A (March 2026, led by Headline) – AI customer-support agents for regulated industries.
• Centrical – $39M Series D (June 2026, led by Leeds Illuminate) – performance intelligence for frontline teams of humans and AI agents.
• Tribal – $10M Seed (May 2026, led by Team8) – a metadata-native platform for deploying enterprise agents.
• Unframe – $50M Series B (May 2026, led by Highland Europe) – managed enterprise AI deployment. Total funding: $100M.
• Nimble – $47M Series B (February 2026, led by Norwest) – real-time web-data infrastructure for AI systems.

Alta ($25M Series A) and Aligned ($60M Series B) – announced just after this window, in early July 2026, and are early signals of the same dynamic continuing into H2.

The funding story, in numbers

Looking at dollars by category tells only part of the story. Across the 24 companies in the core vertical-AI map:

• Total disclosed funding: $716M. The three largest rounds – Tomorrow.io ($210M), NoTraffic ($90M) and Datarails ($70M) – account for $370M, or roughly 52% of all disclosed capital. Vertical AI in Israel is real, but it is highly concentrated at the top.
• Median round size: ~$15M, and 14 of 24 rounds (58%) were seed or Series A – evidence that despite the concentration at the top, there’s genuine early-stage breadth building underneath the mega-rounds.
• Fintech and insurance have the greatest breadth. Nine companies are attacking lending, underwriting, financial planning, private credit and property data independently, rather than clustering around one dominant deal.
• Defence and physical AI attract the largest rounds. More than half of all core vertical-AI capital sits in a single category – battlefield intelligence, autonomy, sensors, weather and traffic infrastructure – reflecting a genuine Israeli domain advantage rather than hype.
• Healthcare is active but still underrepresented. At $59.4M and roughly 8% of disclosed capital, Israel’s healthcare vertical-AI activity is proportionally far behind the US, where Menlo Ventures puts healthcare at close to half of a $3.5B 2025 vertical-AI spending pool.
• The horizontal-agent market is more crowded than vertical AI itself. Functional AI raised nearly as much as two full industry verticals combined, with even tighter concentration among category leaders. A lot of companies describe themselves as “agent platforms.” The harder and more defensible opportunity is owning a complete, regulated industry workflow.

Here are the investors who deployed into Israeli vertical AI in h1 2026 (courtesy of Israelvc.com)

A methodology note: a few companies’ “Israeli” identity runs through a lead investor (TLV Partners backing CareBestie) or an R&D center rather than a founding team based in Israel (Dono, Unframe). They’re included here because Israeli capital and talent are structurally involved, but it’s a judgment call worth flagging rather than asserting as settled fact.

Full funding table

Italicized rows are functional AI, shown for completeness but excluded from the core vertical-AI totals above.

How this connects to the broader thesis

A few frameworks from US-based VCs are worth reading alongside this data:

• Bessemer’s Building Vertical AI playbook argues the winners combine proprietary workflows, multimodal capability, business-model innovation and deep industry distribution.
• Menlo Ventures: Software Finally Gets to Work draws the cleanest conceptual line: vertical SaaS displayed and organised information; vertical AI reasons and executes.
• Menlo’s 2025 State of Generative AI in the Enterprise puts 2025 vertical-AI spending at $3.5B, nearly 3x the year before, with healthcare responsible for roughly $1.5B of it.
• a16z’s Big Ideas 2026 describes the next phase as “multiplayer” vertical AI — agents that coordinate people, other agents and external parties across a full workflow.
• a16z’s enterprise-adoption analysis shows roughly 29% of the Fortune 500 are already paying customers of a leading AI startup, concentrated in coding, support, search and legal.
• Bessemer’s built-world AI roadmap flags construction and real estate — about a quarter of US GDP but the lowest tech spend of any sector — as underinvested.
• Bessemer’s State of Health AI 2026 provides healthcare-specific context, including six health-tech IPOs since 2024 adding $36.6B in market cap.
• Bessemer’s defense-tech roadmap: Five Frontiers for 2026 connects directly to Israel’s strength in defence, autonomy and physical AI.

Going forward, we predict we’ll see more specialised vertical AI startups automating job roles or processes in a specific industry. If you’re an Israeli founder in stealth building in these areas, we’d love to talk to you at Remagine Ventures!

Eze Vidra is the founder of VC Cafe and the co-founder and managing partner of Remagine Ventures, a pre-seed fund investing in ambitious founders at the intersection of AI, technology, entertainment, gaming, and commerce with a spotlight on Israel.

He is a former General Partner at Google Ventures (GV) in Europe, former head of Google for Entrepreneurs in Europe, and founding head of Campus London, Google’s first startup hub. Eze writes on Israeli tech, venture capital, artificial intelligence, and founder strategy.

He is also the founder of Techbikers, a nonprofit that brings together the startup ecosystem on cycling challenges in support of Room to Read.

Latest posts by Eze Vidra (see all)