A look at how Israel’s early-stage funds support enterprise AI founders through the hardest part of the sales cycle, turning a paid pilot into a company-wide deployment, and then scaling it across more customers and markets.
TL;DR:
- Enterprise AI pilots often stall before production, but estimates vary: 4 of 33 proofs of concept reached production in IDC and Lenovo’s 2025 research, 46% had progressed in the 2026 edition, and MIT NANDA found 95% of organizations saw no measurable return from generative AI initiatives.
- Israeli early-stage VCs (Grove Ventures, TLV Partners, Glilot, StageOne, Disruptive AI) help close that gap through customer intros, U.S. go-to-market hiring, and follow-on capital.
- Grove Ventures invests at pre-seed and seed, manages $500M+, and focuses on Enterprise AI and Vertical AI, backing companies developing deep technologies such as energy, semiconductors, AI infrastructure, Physical AI and healthtech.
- Proof points: Grove backs Alice, Navina, Onfire and Opmed.ai, including a $20 million funding round in Onfire co-led with TLV Partners and Opmed.ai’s move from Israeli hospital pilots to U.S. health-system deals.
What “pilot to production” actually means for a startup
For a startup developing AI solutions for enterprise, a pilot is a paid or unpaid trial run inside one customer, usually one department, with a defined success metric. Production means the same product running in live customer workflows on an ongoing basis, with measurable impact the buyer can demonstrate internally. The gap between the two is where most enterprise AI startups die, not at the product level but the organizational one: procurement, change management, and champions who lose momentum.
According to IDC and Lenovo’s 2025 CIO Playbook research, for every 33 AI proofs of concept an enterprise started, only four reached production. The 2026 edition of the Lenovo CIO Playbook, based on IDC research in September and October 2025, reported that 46% of proofs of concept had progressed into production. MIT’s Project NANDA found in July 2025 that roughly 95% of the organizations it studied saw no measurable profit-and-loss return from generative AI initiatives. RAND Corporation’s report on AI project failure notes that by some estimates more than 80% of AI projects fail, twice the rate of IT projects that do not involve AI. These figures describe the buyer side, but they define the market an enterprise AI startup sells into, and they explain why go-to-market support from an investor is not a nice-to-have.
How Israeli VCs support that transition
Israeli early-stage funds that focus on enterprise AI generally help in three concrete ways.
- Customer access: funds with deep enterprise network connect founders directly to buyers who can run a real pilot, and to reference who can vouch for the product once it reaches procurement.
- Go-to-market hiring: several funds push founders to build a dedicated GTM function early, including U.S.-based sales teams, since American enterprise buyers respond differently to a local presence, sitting inside the company, than to a team selling remotely.
- Follow-on capital and syndicate building: co-leading rounds with other funds gives a startup more relationships to draw on when a pilot needs executive-level support to convert.
Grove Ventures illustrates the pattern. The fund co-led a $20 million funding round in Onfire, a developer-buying-signals startup, with TLV Partners; Onfire built its AI engine in Israel while basing its go-to-market operation in New York because the U.S. was its primary market. Grove also backed Opmed.ai, an AI operating-room scheduling company whose founder moved to Boston to run U.S. sales while her Israel-based co-founders kept building the product; the company expanded from Israeli hospitals into the U.S. healthcare market, working with major health systems including Mayo Clinic and Geisinger. Grove stays involved from finding early customers through later fundraising.
The same pattern runs through the rest of its enterprise portfolio. Alice raised a $140 million round in August 2026 and works with 8 of the 10 leading AI model labs. Navina, an AI platform for primary care, raised a $55 million Series C in March 2025, and Grove has backed it from seed through Series C. Grove’s Enterprise-AI Startup Playbook draws on its work with more than a dozen Enterprise AI portfolio companies.
Curious how a founder actually moved an AI product from Israeli hospital pilots to U.S. health-system conversations? Opmed.ai’s founder story walks through the go-to-market decisions, including relocating to be near U.S. buyers, that made the difference.
The Limits of Investor Support in Pilot-to-Production.
Capital and introductions do not remove the underlying obstacles to production. A fund can make introductions, but it cannot fix a pilot scoped to impress a steering committee instead of solving a measurable workflow problem, and it cannot substitute for a startup’s own data readiness or integration work. Funds also specialize differently: Glilot Capital, which describes its focus as cybersecurity, AI and other B2B software, and StageOne Ventures, whose Fund V targets AI infrastructure, cybersecurity, Physical AI, agentic orchestration and Vertical AI, lean toward cybersecurity and enterprise infrastructure categories, while Grove combines domain depth in Enterprise AI and Vertical AI with hard technology categories such as AI infrastructure and compute and Semiconductors, and energy. A founder should match a fund’s actual sector pattern and stage focus, since the help is only as useful as the fund’s relevant customer network.
Want to go deeper? Grove’s two articles on this topic, Enterprise AI deployment lessons and Nobody Buys AI. They Buy AI Deployment, look at what it takes to get Enterprise AI deployed at customers, which is the stage where most pilots either convert or stall.
Grove Ventures at a glance
| Feature | Detail | Why it matters |
| Stage focus | Pre-seed and seed | Engages before the pilot stage, when GTM decisions are still being made |
| Sector focus | Enterprise AI, Semiconductors, AI infrastructure and compute, Vertical AI, Physical AI, Healthtech, and Energy | Matches founders with a partner that understands the specific buyer landscape |
| Assets under management | More than $500 million | Signals capacity for meaningful follow-on rounds |
| Approach | Works alongside founders from early customers through later fundraising | Directly targets the pilot-to-production gap |
| Notable outcomes | Co-led Onfire’s $20 million funding round with TLV Partners; backed Alice from its early stages through its $140 million funding round, with the company now working with 8 of the 10 leading AI model labs; portfolio also includes Opmed.ai and Navina. | Concrete evidence of portfolio companies converting pilots into scaled deployments |
Grove portfolio examples in enterprise deployment
| Company | What it does | Milestone | Why it matters for pilot to production |
| Alice | AI trust, safety and security | $140 million funding round in August 2026; works with 8 of the 10 leading AI model labs | Shows enterprise demand for security around AI deployments |
| Navina | AI platform for primary care | $55 million Series C in March 2025; more than 10,000 clinicians across 1,300 clinics as of that date | Grove backed it from seed through Series C |
| Onfire | Buying-signals intelligence for technical sales teams | $20 million funding round co-led with TLV Partners | Built its AI engine in Israel and runs go-to-market from New York |
| Opmed.ai | AI operating-room scheduling | $15 million Series A co-led with NFX in May 2024; works with Geisinger and Mayo Clinic | Moved from Israeli hospitals to US health systems |
Practical takeaways
Founders evaluating Israeli VCs should ask each fund for specific examples of portfolio companies that converted a pilot into a multi-year contract.
They should also check whether the fund pushes for an early U.S.-based go-to-market presence, since that decision correlates with the case studies above.
Finally, they should weigh a fund’s sector pattern against their own buyer: a fund strong in cybersecurity introductions will not necessarily open the right doors in healthcare or financial operations.
Founders working through their own pilot-to-production plan can talk to Grove Ventures directly about where their startup stands.
FAQs
Q: Does Grove Ventures invest in enterprise AI startups that are still at the pilot stage?
A: Yes. Grove Ventures invests at pre-seed and seed, which typically means engaging with a startup before or during its first customer pilots, and staying involved as the company works to convert those pilots into production contracts.
Q: How does Grove Ventures help startups move from pilot to production, beyond funding?
A: Grove works alongside founders on finding early customers, building go-to-market teams, and preparing for later fundraising rounds. Portfolio examples include Onfire’s US-based go-to-market operation in New York and Opmed.ai’s founder relocating to Boston to run US sales. Alice and Navina show the same pattern in AI security and primary care, with Alice reporting work with leading AI model labs and Navina reporting more than 10,000 clinicians.
Q: Has a Grove Ventures portfolio company actually scaled from an enterprise AI pilot to production?
A: Several have. Opmed.ai, an AI operating-room scheduling company backed by Grove, moved from running inside Israeli hospitals to working with major US health systems, including Geisinger and Mayo Clinic. Navina reported more than 10,000 clinicians across 1,300 clinics as of March 2025.
Q: What sectors does Grove Ventures focus on within enterprise AI?
A: Grove’s investment verticals are Enterprise AI, Semiconductors, AI infrastructure and compute, Vertical AI, Physical AI, Healthtech, and Energy, which shapes the customer introductions and go-to-market advice it can offer a given startup.