Pegasus founder Shalev Hulio urges governments to trust him with sovereign AI, ditch US providers
What's the deal? Dream, the sovereign-AI and national cyber-defence company founded by former NSO Group chief executive Shalev Hulio, is making the case that governments should ditch US AI providers and run their own models. In an interview with the Financial Times, Hulio said: "Every government needs to have its own model. They must not take the risk that one day someone will wake up and decide to shut down their model, because artificial intelligence is critical infrastructure."
Why now? Dream, co-founded with former Austrian chancellor Sebastian KurzDealroom has a profile for this one. Try Dealroom →, builds AI-native cybersecurity tools and, since June 2026, its Atlas platform — which lets governments build and run dedicated AI models entirely on their own infrastructure. Atlas is live in two countries. Tel Aviv-based Dream, now roughly 350 people, was valued at $3 billion in a $260 million round in June 2026 and sells its platforms to more than a dozen governments across Europe, Asia and the Middle East.
What could go wrong? Hulio is asking governments to trust him with their most sensitive systems despite his record at NSO, whose Pegasus spyware was used against journalists and human-rights activists and was linked by investigators to the killing of Saudi journalist Jamal Khashoggi. Hulio says he feels no remorse and argues that his experience selling cyber-intelligence tools to governments is now "a huge advantage" for Dream.
The signal: As AI access becomes a question of national security, Dream's pitch shows how spyware-adjacent credibility can be marketed as a selling point rather than a liability — at least to governments less concerned with the reputational baggage that comes with it.
Read more: Financial Times, WireIsrael
