Modelyo
Sovereign Confidential Cloud platform for sensitive data, applications, and AI workloads.
Website: https://modelyo.com/
Cover Block
From the public record
| Field | Value |
|---|---|
| Name | Modelyo |
| Tagline | Sovereign Confidential Cloud platform for sensitive data, applications, and AI workloads [Israel Hayom, October 2026] |
| Headquarters | Tel Aviv District, Israel [LinkedIn, retrieved 2026] |
| Founded | 2024 [Crunchbase] |
| Stage | Seed [Israel Hayom, October 2026] |
| Business Model | SaaS |
| Industry | Security |
| Technology Type | Software (Non-AI) |
| Geography | Middle East / North Africa |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+): Benny Meir, Dr. Itzhak Aviv, Dr. Michael Kiperberg [Israel Hayom, October 2026] [Calcalist Tech, October 2026] |
| Funding Label | Seed |
| Total Disclosed | Approximately $17,000,000 [Israel Hayom, October 2026] |
Links
From the public record
- Website: https://modelyo.com/
The Short Version
PUBLIC Modelyo is building a sovereign confidential cloud platform for enterprises and public-sector operators that need to run sensitive data, applications, and AI workloads on third-party cloud infrastructure without surrendering control of the data, encryption keys, or model weights, and the company merits attention now because it disclosed an approximately $17 million seed round in October 2026 led by OlivTree [Israel Hayom, October 2026] [Calcalist Tech, October 2026]. Founded in 2024 in Israel, the company is pitching itself at the intersection of confidential computing and data sovereignty, with public reporting describing a product that protects information while it is actively being processed rather than only when stored or transmitted, a narrower and more technically consequential claim than standard cloud-security positioning [Calcalist Tech, June 2026] [Dealroom News].
The commercial thesis is straightforward even if public proof points remain limited: if Modelyo can let regulated and security-sensitive organizations use public cloud and AI infrastructure while keeping providers blind to underlying data contents and keys, it addresses a real procurement bottleneck across government, defense, banking, healthcare, pharmaceuticals, and critical infrastructure [Israel Hayom, October 2026] [Dealroom News]. Public sources also point to more than 20 patents and patent applications and to approval for use cases tied to the Israeli Government Nimbus environment, although those claims rely on a mix of company-adjacent and LinkedIn-sourced evidence and should be treated as promising rather than fully corroborated [Dealroom News] [LinkedIn, retrieved 2026].
The founding bench is a meaningful part of the story. CEO Benny Meir is described as a reserve lieutenant colonel with more than two decades in special operations and technology capability building, while co-founders Dr. Itzhak Aviv and Dr. Michael Kiperberg bring backgrounds in defense communications, encryption, secure virtualization, and computing infrastructure, which is directly relevant to a product centered on data-in-use protection and trusted execution environments [Israel Hayom, October 2026] [Calcalist Tech, October 2026] [Crunchbase].
On the company-building side, the disclosed profile is still early. Modelyo operates as a SaaS business at seed stage, had a reported 30-person workforce as of October 2026, and appears to have been founder-funded before the October 2026 financing; public materials do not establish revenue, named customers, or deployment scale, so the next 12 to 18 months likely turn on converting technical credibility into referenceable production adoption outside Israel [Calcalist Tech, October 2026] [Israel Hayom, October 2026].
Single-source, plausible -- Core company, funding, and founder facts are corroborated by Israel Hayom, Calcalist Tech, and Crunchbase; several product-adoption and approval claims rely on single-source or LinkedIn-based evidence.
Taxonomy Snapshot
| Axis | Value |
|---|---|
| Stage | Seed |
| Business Model | SaaS |
| Industry / Vertical | Security |
| Technology Type | Software (Non-AI) |
| Geography | Middle East / North Africa |
| Growth Profile | Venture Scale |
| Founding Team | Co-Founders (3+) |
| Funding | Seed, total disclosed approximately $17,000,000 [Israel Hayom, October 2026] |
The Company in Brief
PUBLIC
Modelyo appears to be a very young Israeli security company, and the public record is still thin enough that chronology matters more than category language. Crunchbase lists the company as founded in 2024, while the company presents itself as a provider of a "Sovereign Confidential Cloud" platform built for sensitive data, applications, and AI workloads [Crunchbase] [Modelyo]. The headquarters location in the available structured record is Tel Aviv District, Israel, which is directionally consistent with the founders' LinkedIn geography but not independently established through a filing in the materials provided, so the cleaner public fact is that Modelyo is operating from Israel [Crunchbase] [Modelyo].
The company website frames the product around running sensitive workloads on existing cloud infrastructure while keeping control of data and related security controls with the customer, rather than handing that control to the underlying cloud provider [Modelyo]. That positioning is also reflected in Crunchbase's company profile, which categorizes Modelyo as a cybersecurity software company at seed stage [Crunchbase].
The first clear milestone in the public record is the company's 2024 founding entry on Crunchbase, followed by the buildout of its public web presence and company blog, both of which indicate an active go-to-market and thought-leadership posture by 2026 [Crunchbase] [Modelyo]. Crunchbase also identifies co-founders Michael Kiperberg and Itzhak Aviv in operating roles at the company, while the website presents the business under the Modelyo name and its confidential-cloud positioning [Crunchbase] [Modelyo]. Within the source set allowed for this section, no legal entity filing, incorporation number, or prior financing history is confirmed beyond the seed-stage company classification shown on Crunchbase [Crunchbase].
Single-source, plausible -- This section relies on Crunchbase and the company website; key company descriptors are corroborated across both, but legal entity detail and fuller milestone history are not confirmed in the provided filing-level sources.
What They Have Built
MIXED Product and Technology
Modelyo is positioning around a narrow but consequential problem: how to use public cloud and AI infrastructure without surrendering control of the underlying data. Public descriptions of the product are consistent on the core claim. The company says its "Sovereign Confidential Cloud" is built so governments, banks, defense organizations, healthcare providers, pharmaceutical companies, critical-infrastructure operators, and other enterprises can run sensitive data, applications, and AI workloads on cloud infrastructure while retaining control of data, encryption keys, and model weights [Israel Hayom, October 2026] [Calcalist Tech, October 2026].
The technical wedge, as described in press and third-party commentary, is confidential computing and data-in-use protection rather than conventional encryption at rest or in transit. Calcalist Tech reported in June 2026 that Modelyo focuses on protecting information while it is being processed [Calcalist Tech, June 2026]. Dealroom News similarly described the platform as running workloads on third-party cloud infrastructure without giving cloud providers access to data contents or keys [Dealroom News]. External writeups from Paperclip and Duality frame the same concept more explicitly, stating that Modelyo aims to keep data encrypted or isolated during active processing, including exposure points in RAM, CPUs, and processing servers [Paperclip] [Duality]. Those descriptions line up directionally, but they stop short of a verified public demo with architectural detail, performance benchmarks, or deployment evidence.
The public record also suggests Modelyo is trying to package this capability for highly regulated and classified environments, not only for standard enterprise security buyers. A LinkedIn post cited in the research says the company has been approved for Roved 5, or Layer 5, for the Israeli Government Nimbus Cloud environment, which if accurate would matter because it implies alignment with government-grade handling requirements [Yehonatan Kfir, PhD - Wisery Labs | LinkedIn, retrieved 2026]. Separately, the company website says enterprise customers receive a named technical contact and 24/7 support for production issues [Modelyo]. That support claim is company-sourced, and the broader product narrative still lacks public customer case studies, verified integrations, or independent technical validation. The result is a product story that is coherent and timely, with evidence strongest on problem framing and weakest on externally verified implementation depth.
Single-source, plausible -- Core product positioning is corroborated by Israel Hayom and Calcalist Tech, while several technical and compliance-related claims rely on single-source press summaries, LinkedIn posts, or company materials.
Market Size and Demand
PUBLIC The market matters now because Modelyo is selling into a live contradiction: enterprises want public-cloud scale and AI adoption, but a meaningful share of regulated workloads still cannot move unless data remains controlled during processing, not only at rest or in transit [Calcalist Tech, June 2026] [Israel Hayom, October 2026].
Public reporting on Modelyo does not provide a sourced TAM, SAM, or SOM, and the available materials do not cite a third-party market study specific to sovereign confidential cloud infrastructure [Israel Hayom, October 2026] [Calcalist Tech, October 2026]. What the evidence does show is the demand contour. Modelyo is explicitly positioned at governments, banks, defense organizations, healthcare providers, pharmaceutical companies, critical-infrastructure operators, and other enterprises handling sensitive workloads, a customer set that tends to buy on compliance, resilience, and control rather than on pure infrastructure cost [Israel Hayom, October 2026]. That matters because these are large IT spenders already using cloud and increasingly experimenting with AI, yet public sources still frame data-control limits as the gating issue for broader deployment [Calcalist Tech, June 2026] [Calcalist Tech, October 2026].
The near-term demand drivers are clearer than the market size. Modelyo's pitch rests on confidential computing and data sovereignty: keeping cloud providers from accessing data contents or encryption keys, and protecting information while it is being processed [Calcalist Tech, June 2026] [Dealroom News]. That lines up with a wider enterprise shift from protecting stored data to protecting data in use, especially as AI workloads require large-scale processing of proprietary records, operational data, and potentially classified or regulated information [Paperclip, 2026] [Duality, 2026]. The cited public coverage also places Modelyo's expansion effort in international markets, which suggests management sees this as a cross-border policy problem as much as a technical one, though the current evidence does not quantify commercial traction by region [Israel Hayom, October 2026] [Calcalist Tech, October 2026].
Adjacent and substitute markets are easier to define than direct market size. At one end, Modelyo overlaps with broader cloud security, key management, and encryption-control budgets, since its value proposition is that customers retain control of data, keys, and model weights even on third-party infrastructure [Israel Hayom, October 2026]. At the other end, it touches confidential computing, sovereign cloud, and secure AI infrastructure categories, which are often budgeted separately inside government and heavily regulated enterprises when public-cloud adoption hits policy constraints [Dealroom News] [Duality, 2026]. The practical substitute is not only another startup. In many cases it is the status quo of keeping sensitive workloads on isolated on-premise or classified systems, delaying AI deployment, or limiting which datasets can reach external clouds at all [Calcalist Tech, June 2026] [Paperclip, 2026].
Regulation and public-sector procurement appear to be the strongest macro tailwinds in the available evidence. A cited LinkedIn post states that Modelyo had approval for "Roved 5 (Layer 5)" in connection with classified data processing on the Israeli Government Nimbus Cloud, which, if borne out in deployment, would place the company inside one of the harder segments of cloud adoption rather than at the edge of discretionary enterprise tooling [Yehonatan Kfir, PhD - Wisery Labs | LinkedIn, retrieved 2026]. That claim is not independently confirmed in mainstream reporting, so it should be treated carefully, but it is directionally consistent with the company's stated focus on sovereign control for public-sector and defense-grade use cases [Israel Hayom, October 2026]. More broadly, macro conditions favor vendors that can reduce the tradeoff between AI adoption and data exposure, especially where procurement requirements increasingly ask where data sits, who controls keys, and whether processing can occur without provider visibility [Calcalist Tech, June 2026] [Calcalist Tech, October 2026].
| Cited market signal | What the source supports | Relevance to Modelyo |
|---|---|---|
| Regulated-sector targeting [Israel Hayom, October 2026] | Governments, banks, defense, healthcare, pharma, and critical infrastructure are named target users | Indicates a budget pool tied to compliance and national-security constraints rather than broad SMB demand |
| Data-in-use protection [Calcalist Tech, June 2026] | The problem is framed as protecting information during processing and retaining control of encryption keys | Suggests demand is tied to AI and cloud adoption where conventional encryption is insufficient |
| Third-party cloud without provider access [Dealroom News] | Workloads can run on external cloud infrastructure without giving providers access to data contents or keys | Points to a spend category adjacent to confidential computing and sovereign cloud controls |
| Public-sector cloud eligibility claim [Yehonatan Kfir, PhD - Wisery Labs | LinkedIn, retrieved 2026] | Reported approval related to classified processing on Israeli Government Nimbus Cloud |
The public evidence does not size the market, but it does show where the budget should come from: regulated cloud migration, secure AI deployment, and sovereign-compute mandates. That is a narrower but potentially more durable entry point than generic cloud security, provided the company can convert policy relevance into verified deployments.
Single-source, plausible -- Section relies on named public reporting and company-adjacent sources for demand drivers, but no independently sourced third-party market sizing was available.
Who Else Is Fighting for This
MIXED
Modelyo appears to be positioning itself less against a single named startup peer and more against the default architecture choice enterprises already use, conventional public cloud for sensitive workloads, plus point solutions in encryption and secure infrastructure [Israel Hayom, October 2026] [Calcalist Tech, June 2026].
The public record is thin on named head-to-head competitors, which is itself a useful signal. Coverage consistently describes Modelyo through the problem it addresses, running sensitive data, applications, and AI workloads on third-party cloud infrastructure while keeping control of data, keys, and model weights, rather than through direct comparison with another vendor class [Israel Hayom, October 2026] [Dealroom News]. That places it in a crowded but still forming segment around confidential computing, data sovereignty, and secure cloud execution, where the practical alternatives for a buyer are likely to split three ways: incumbent hyperscaler-native security controls, specialist confidential-computing infrastructure, and the status quo of keeping the most sensitive workloads off shared cloud entirely. Only the first and third of those alternatives are directly evidenced in the source set, so any finer segmentation should be treated as directional rather than established by public reporting [Calcalist Tech, June 2026] [Modelyo].
Modelyo's clearest edge today is fit between product claim and founding profile. The company is selling into sectors where buyer trust often tracks technical credibility and operational familiarity, and its founders are publicly described with backgrounds in special operations, defense communications, encryption systems, secure virtualization, and computing infrastructure [Israel Hayom, October 2026] [Calcalist Tech, October 2026]. There is also a regulatory wedge in the Israeli public sector narrative: a LinkedIn post cited in the research says Modelyo was approved for Roved 5 on the Israeli Government Nimbus Cloud, which, if borne out in customer procurement, could matter more than broad brand awareness in the near term [LinkedIn, retrieved 2026]. That edge looks durable only if the company converts technical trust into referenceable deployments and procurement entrenchment. Without that conversion, the advantage is perishable, because hyperscalers and larger security vendors already own the cloud relationship and can bundle adjacent controls into existing contracts [Modelyo] [Calcalist Tech, October 2026].
The company is most exposed where platform ownership and distribution matter more than architecture. Public sources say Modelyo runs on third-party cloud infrastructure without exposing data contents or keys to the provider, which is compelling as a control layer, but it also means the company does not own the underlying compute channel [Dealroom News]. Amazon Web Services and Google Cloud are not named as competitors in the source set, but the Nimbus reference effectively underscores the point that major cloud vendors remain the infrastructure gatekeepers in at least one strategic market context [LinkedIn, retrieved 2026]. If a buyer decides native cloud controls are sufficient, or prefers to concentrate spend with an existing cloud provider rather than add a specialist layer, Modelyo can lose before a product bake-off begins. The absence of named customers, partner announcements, or commercial deployment metrics in the available reporting makes that go-to-market risk hard to discount [Calcalist Tech, October 2026] [Israel Hayom, October 2026].
The most plausible 18-month scenario is a split market rather than a winner-take-all outcome. Modelyo is the likely winner if regulated buyers increasingly decide that AI adoption requires a separate sovereignty and confidential-execution layer, especially in government, defense, banking, and critical infrastructure settings where control of keys and workloads carries procurement weight beyond raw feature breadth [Israel Hayom, October 2026] [Paperclip]. The likely loser in that scenario is the status quo of keeping sensitive workloads off public cloud altogether, because Modelyo's pitch is explicitly to bridge classified or mission-critical environments with cloud infrastructure rather than force an on-premises tradeoff [Calcalist Tech, June 2026] [Duality]. The reverse scenario is also straightforward: if native cloud security controls satisfy regulators and enterprise security teams, the specialist vendor with the least owned distribution, here Modelyo, is the one under the most pressure.
Opportunity
PUBLIC
The prize here is not another security tool but a control layer for regulated cloud adoption, and if Modelyo executes, it could become the infrastructure that lets governments and large enterprises move sensitive workloads, including AI, onto third-party cloud without surrendering control of data, keys, or model weights [Israel Hayom, October 2026] [Calcalist Tech, October 2026].
The headline opportunity is straightforward. Modelyo is targeting a real friction point in enterprise computing: organizations want public-cloud scale and AI capability, but some of the most valuable workloads cannot move unless data remains protected during processing, not only at rest or in transit [Calcalist Tech, June 2026] [Paperclip, 2026] [Duality, 2026]. That is why the company is positioning its platform around confidential computing and data sovereignty across government, banking, defense, healthcare, pharma, and critical infrastructure, all sectors where the buyer penalty for a security or control failure is unusually high [Israel Hayom, October 2026]. The outcome that matters is not consumer-scale ubiquity. It is becoming a default trust layer for a narrower set of high-value, compliance-bound cloud and AI deployments.
That outcome looks reachable, at least in outline, because the early ingredients line up. The company raised about $17 million in seed financing in October 2026, led by OlivTree, to accelerate R&D and international commercial expansion [Israel Hayom, October 2026] [Calcalist Tech, October 2026]. Public reporting also places the team at 30 employees as of October 2026, with technical leadership backgrounds in encryption, secure virtualization, and defense computing, which is the sort of founder-market fit that matters more in this category than brand polish or broad awareness [Calcalist Tech, October 2026] [Israel Hayom, October 2026].
| Scenario | What happens | Catalyst | Why it's plausible |
|---|---|---|---|
| Sovereign cloud standard | Modelyo becomes a preferred control layer for public-sector and defense-adjacent workloads that must run on approved cloud infrastructure without exposing plaintext data or keys | Regulatory acceptance of classified or highly sensitive processing on approved cloud environments, including the reported Roved 5 approval linked to the Israeli Government Nimbus context [Yehonatan Kfir, PhD - Wisery Labs | LinkedIn, retrieved 2026] | The company is explicitly built for governments and defense-sensitive users, and public sources describe its product as enabling sensitive workloads on third-party cloud while preserving control over data and keys [Israel Hayom, October 2026] [Dealroom News] |
| AI control plane for regulated enterprises | Modelyo wins budgets as enterprises move AI from pilots to production and need technical guarantees around data-in-use security | A step-change in enterprise AI deployment inside banking, healthcare, pharma, and critical infrastructure, where model usage expands faster than internal security policies [Calcalist Tech, June 2026] [Israel Hayom, October 2026] | Multiple sources tie the product directly to AI workloads and to the problem of protecting information while processing, which is one of the harder blockers for sensitive AI use cases [Calcalist Tech, June 2026] [Calcalist Tech, October 2026] |
| Cross-border sovereign platform | Modelyo turns sovereignty requirements into an exportable platform for markets with strict data-residency or national-security constraints | Successful commercial entry into overseas markets after the seed round, which public reporting says is intended to fund international expansion [Calcalist Tech, October 2026] [Israel Hayom, October 2026] | The product message is not country-specific. It maps to a broader class of buyers that want public cloud economics without ceding operational control of sensitive information [Israel Hayom, October 2026] [Dealroom News] |
The scenarios point to the same core idea. Modelyo does not need to win every enterprise security budget. It needs to become essential in a small number of expensive, slow-moving, high-trust environments where losing control of data rules out cloud adoption altogether.
What compounding looks like in this model is a credibility flywheel. One successful deployment in a regulated or government setting can reduce proof burdens for the next buyer, because these customers often ask the same questions about key custody, data exposure during processing, and cloud-provider visibility [Calcalist Tech, June 2026] [Israel Hayom, October 2026]. The product itself also appears suited to that motion: if it can let customers run on existing third-party cloud infrastructure rather than force a separate stack, adoption can piggyback on budgets that already exist for cloud and AI, while support intensity and named technical coverage suggest a high-touch enterprise sale designed for long-lived production use [Dealroom News] [Modelyo].
There is a second compounding layer if the company can turn technical acceptance into policy acceptance. The reported patent portfolio of more than 20 patents and patent applications does not prove defensibility on its own, but paired with founder expertise in encryption and secure virtualization, it suggests the company is trying to own a specific architectural layer rather than sell a transient feature set [Dealroom News] [Calcalist Tech, October 2026]. In this category, distribution can come less from developer virality and more from trust accumulation, referenceability, and inclusion in the shortlist for workloads that cannot otherwise move to cloud.
The size of the win is easiest to frame as strategic infrastructure value rather than near-term revenue arithmetic. No public market-sizing figure or direct public comparable is confirmed in the available sources, so a precise valuation case would overstate what the evidence supports. Even so, if the "AI control plane for regulated enterprises" scenario plays out, the company could plausibly become an acquisition target or large independent platform in security and cloud infrastructure, because it sits at the intersection of sovereign cloud, confidential computing, and enterprise AI governance, three budgets that are converging in the public reporting around its product thesis [Calcalist Tech, October 2026] [Israel Hayom, October 2026]. A multibillion-dollar outcome is possible here (scenario, not a forecast), but it depends on proving that Modelyo is not only technically credible, but operationally trusted enough to become part of how sensitive institutions adopt cloud and AI at all.
Single-source, plausible -- Built primarily from Israel Hayom, Calcalist Tech, Dealroom News, and company materials; the upside framing is analytical and several technical and regulatory signals rely on single-source or company-adjacent evidence.
Sources
From the public record
[Israel Hayom, October 2026] Modelyo raises approximately $17 million to secure next generation of cloud, AI | https://www.israelhayom.com/2026/10/22/modelyo-raises-approximately-17-million-to-secure-next-generation-of-cloud-ai
[LinkedIn, retrieved 2026] Benny Meir - modelyo | LinkedIn | https://www.linkedin.com/in/benny-meir/
[Crunchbase] Modelyo - Crunchbase Company Profile & Funding | https://www.crunchbase.com/organization/modelyo
[Calcalist Tech, October 2026] Modelyo raises $17 million to let companies use AI without exposing their secrets | https://www.calcalistech.com/ctechnews/article/rju100ow2ml?ref=thesaasnews.com
[Calcalist Tech, June 2026] Data is the fuel of the new world. We must make sure it remains ours. | https://www.calcalistech.com/ctechnews/article/5d1bf7tei
[Dealroom News] Modelyo raises $17M to run AI in the cloud without exposing company secrets | Dealroom News | https://dealroom.co/news/165044-modelyo-raises-17m-to-run-ai-in-the-cloud-without-exposing-company-secrets/
[Modelyo] Modelyo | https://modelyo.com/
[Paperclip, 2026] Modelyo - Sovereign Confidential Cloud: Protecting Data in Use | https://www.paperclip.com/blog/modelyo-sovereign-confidential-cloud-protecting-data-in-use
[Duality, 2026] Sensitive data, in use, stays protected: Modelyo secures cloud and AI workloads | https://dualitytech.com/blog/modelyo-secures-cloud-and-ai-workloads/
[WordPress.tv, August 2026] Keynote - Fireside Chat | https://wordpress.tv/2026/08/20/keynote-fireside-chat-2/
Articles about Modelyo
- Modelyo's Sovereign Cloud Protects Data in Use for Israeli Defense and Banks — The Israeli startup, founded by defense and encryption veterans, raised $17 million to take its confidential computing platform to international markets.