Dream raises $260M to build sovereign AI for governments

Key Takeaways

- Dream raised $260 million led by Group 11 and Bicycle Capital, tripling its valuation to $3 billion in under a year.
- The company provides governments with sovereign AI and cyber defense tools to avoid reliance on US or Chinese tech providers.
- The White House ban on foreign nationals using Anthropic's newest AI models has accelerated demand for nationally controlled AI infrastructure.
Dream, the Tel Aviv-based cybersecurity startup building AI tools for governments, has raised $260 million at a $3 billion valuation. The round, led by Group 11 and Bicycle Capital, nearly triples the company's worth from its $1 billion valuation after a $100 million Series B last year. The capital will fund deployments across Europe, the Middle East, Asia, and the Americas.
Co-founder Sebastian Kurz, the former Austrian chancellor, told Bloomberg that European nations face urgent pressure to prepare for emerging cyber threats. His co-founder Shalev Hulio, previously CEO of the controversial NSO Group, framed the company's mission in starker terms: governments can either depend on AI systems controlled by the US or China, or build capabilities they fully own.
Dream was founded to eliminate that trade-off, according to Hulio. The company has reportedly secured over $300 million in total contract value since beginning commercial operations in late 2024, with estimated annual recurring revenue now exceeding $100 million.
Why sovereign AI matters now
The timing of Dream's raise isn't accidental. Last week, the White House banned foreign nationals from accessing Anthropic's newest AI models. For Hulio, this confirmed what he's been arguing: nations that rely on foreign AI clouds and tools face real risk of being cut off when geopolitics shift.
In a LinkedIn post reacting to the Anthropic ban, Hulio predicted the decision would "accelerate one of the biggest shifts in modern history." He compared AI infrastructure to financial and defense systems, arguing no government would accept foreign control over those either. "Smart governments no longer see AI as software," Hulio wrote. "They see it as critical infrastructure."
What Dream actually sells
Dream's flagship product is Atlas, a platform that lets governments deploy AI agents and large language models inside secure, self-contained environments. The company is also launching a custom AI platform for state-owned enterprises seeking greater control over their data. The pitch is independence: run capable AI without data leaving national borders or depending on providers who might restrict access.
This differs from typical enterprise security vendors. Dream isn't selling endpoint protection or threat detection. It's selling the underlying infrastructure that lets a government operate AI capabilities the same way it operates power grids or telecommunications networks.
The controversial founding team
Dream's leadership makes this raise notable beyond the numbers. Hulio ran NSO Group, the Israeli firm behind Pegasus spyware used by governments to surveil journalists and dissidents. That history has drawn criticism, with some questioning whether he should be building tools for state actors.
Kurz, meanwhile, resigned as Austrian chancellor in 2021 amid a corruption investigation. He denied wrongdoing but left politics. The pairing of these two figures selling security tools to governments has generated polarized reactions online, with Hacker News discussions split between concerns about the founders' backgrounds and acknowledgment that the underlying problem, foreign AI dependence, is real.

Where the market is heading
Dream isn't alone in betting on sovereign AI. European governments have been vocal about reducing dependence on American tech giants, and the Anthropic ban adds urgency. The company's LinkedIn announcement put it bluntly: "AI is the next layer" of national infrastructure, following roads, grids, and defense systems. "This round is about making sure they can own that one too, turning nations into super nations."
Whether governments will actually pay for this remains the test. Dream claims $300 million in contracts and $100 million in ARR, but those figures are estimates and the company hasn't disclosed specific customers. Government sales cycles are notoriously slow and political. A change in administration can kill a deal years in the making.
The $3 billion question
At a $3 billion valuation with around $100 million in ARR, Dream trades at roughly 30x revenue. That's aggressive for a company selling to governments, where growth can be lumpy and contract renewals uncertain. The bet is that sovereign AI becomes non-negotiable for major nations, and Dream gets there first.
Hulio's LinkedIn post closed with a prediction: "This isn't the end of a story. It's the beginning of a global race toward sovereign AI." If he's right, Dream's current valuation might look cheap in retrospect. If sovereign AI turns out to be more rhetoric than budget line, the company will need to find other buyers.
Frequently Asked Questions
What is sovereign AI?
Sovereign AI refers to artificial intelligence infrastructure owned and operated within a nation's borders, without dependence on foreign cloud providers or technology vendors. It gives governments control over their AI capabilities and data.
Who founded Dream?
Dream was founded in 2023 by Shalev Hulio, former CEO of NSO Group, and Sebastian Kurz, former Chancellor of Austria.
How much has Dream raised in total?
Dream has raised at least $360 million, including $100 million in its 2025 Series B and $260 million in its latest round announced June 2026.
Why did the Anthropic ban matter for Dream?
The White House ban on foreign nationals using Anthropic's newest models demonstrated that nations relying on US AI providers can lose access due to policy changes, validating Dream's pitch for nationally controlled AI infrastructure.
What is Dream's Atlas platform?
Atlas is Dream's flagship product that allows governments to deploy AI agents and large language models in secure, self-contained environments without data leaving national borders.
Logicity's Take
The Anthropic ban handed Dream a perfect sales pitch at the perfect time. But the company's real challenge isn't convincing governments that sovereign AI matters. It's convincing them to write checks before the next election cycle. Government procurement is where startups go to age. Dream's 30x revenue multiple assumes it can close deals at enterprise speed while selling to bureaucracies. That tension, not the technology, will determine whether this bet pays off.
Need Help Implementing This?
If you're exploring AI infrastructure for your organization or evaluating sovereign deployment options, Logicity's research team can help you navigate vendor options and implementation strategies. Contact us for tailored analysis.
Source: PYMNTS | / PYMNTS
Huma Shazia
Senior AI & Tech Writer
Related Articles
Browse all
Indian Startup IPOs 2026: ₹47,000 Cr Pipeline Reshapes Exit Strategy
India's startup IPO market is set for its biggest year yet, with unicorns like Flipkart, Zepto, and OYO planning to raise over ₹47,000 Cr. But public market investors are demanding profitability over hype, forcing founders to rethink their approach to going public.

Griffin Retreat 2026: $200Bn Founder Network Lessons
Inc42's Griffin Retreat gathered 100 founders representing $200 billion in valuation for closed-door strategy sessions. Here's what business leaders can learn from India's most exclusive founder gathering about building networks, scaling companies, and navigating the shift from growth to institutional building.

Loopworm Insect Protein: How This Bengaluru Startup Cuts Biologics Costs by 80%
Loopworm is using silkworms and black soldier flies to manufacture protein biologics and animal feed ingredients, potentially slashing pharmaceutical manufacturing costs to one-fifth of traditional methods. The startup reported ₹4.5 Cr revenue in FY25 and projects ₹15-18 Cr for FY26.

TraqCheck $8 Million Series A: AI Recruitment Agents Nina and Trace Target European Expansion
Indian enterprise tech startup TraqCheck just closed an $8 million Series A round to scale its AI-powered recruitment agents across Europe. The company, which started with background verification, is now building autonomous agents that source candidates, initiate conversations, and connect them directly with hiring managers.


