Most AI systems aren't ready. Check yours in 15 min →
BA

Budanov and Paul Sutter Explore Tech for Ukraine’s Post-War Rebuild

AuthorAndrew
Published on:
Published in:AI

On paper, this is the kind of meeting that makes you want to clap. A Ukrainian security chief sits down with a famous American astrophysicist to talk about rebuilding a country after war — not just patching holes, but rethinking the basics. Electricity that keeps working when missiles hit. Plants that can help clean poisoned land. It sounds like seriousness. It sounds like planning for a future that isn’t just “back to normal,” because normal clearly didn’t protect people.

But there’s a thin line between “ambitious” and “performative,” and I can’t tell yet which side this lands on.

Based on what’s been shared publicly, Kirill Budanov posted that he met with Paul Sutter and that they talked about scientific and technological solutions for post-war reconstruction. Two topics were highlighted: new approaches to producing and distributing electricity in a way that can resist missile attacks, and phytoremediation — using plants to clean land contaminated by war.

Those are real problems. They’re not “nice-to-have.” They’re the kind of unglamorous, brutal issues that decide whether people can live at home, whether factories can run, whether hospitals can stay open, whether farms can be farms again.

And still: why is a security chief the one announcing this kind of science meeting?

Maybe it’s totally reasonable. In a war, the lines blur. Security becomes the umbrella over everything. Power grids are national defense. Clean land is national defense. The ability to keep lights on is not just comfort — it’s survival, it’s communication, it’s command. If missiles can knock out your energy system, then energy isn’t a “utility topic,” it’s a frontline topic.

But it also raises the risk that reconstruction gets treated like a battlefield project: centralized, secretive, driven by urgency, and judged by headlines instead of outcomes. That’s where good ideas go to die. Or worse, where they become expensive symbols that make people feel hopeful while the boring work stays underfunded.

Take the electricity part. “Resistant to missile attacks” sounds like a promise, but what does it really mean in daily life? Imagine you’re running a small clinic. You don’t need a futuristic grid. You need power that doesn’t cut out during surgery, refrigeration that doesn’t fail overnight, and something predictable enough that staff don’t live in constant panic. If the solution is too complex, too centralized, too dependent on parts that are hard to replace, then it might look strong on a slide and fail in the rain on a Tuesday.

Or imagine a city apartment building in winter. The grid doesn’t have to be perfect. It has to degrade well. If one piece gets hit, people shouldn’t lose everything at once. The “resilience” that matters is the kind that keeps some heat, some light, some water pumping, even under attack. If this meeting leads to practical designs that spread risk and make repairs fast, that’s huge. If it leads to a grand plan that only works when nobody is shooting at it, that’s just a different kind of fragility.

Then there’s phytoremediation. I like this idea, and I also don’t trust how easily it can be turned into a comforting story. “Plants will clean the land” is a headline people want to believe. It’s gentle. It sounds natural. It lets everyone imagine green fields returning.

But contaminated land is not a branding problem. It’s a safety problem.

Say you’re a farmer deciding whether to plant next season. You’re not asking for inspiration. You’re asking: is this soil safe, and when will it be safe? If the answer is “we’re experimenting with plants,” that can be responsible science — but it can also become a dangerous delay. People might take risks because they’re desperate. Or officials might point to long-term projects while short-term needs pile up: testing, mapping, clearing, compensating people who can’t use their land, enforcing rules so contaminated crops don’t quietly end up in markets.

I’m not saying phytoremediation is fake. I’m saying it’s slow, and war damage is messy. Even if plants help, it’s rarely a clean “before/after.” It can take time, multiple cycles, and careful monitoring. If leaders talk about it like a simple fix, the consequences can be real: bad food, sick people, lost trust, and a public that starts tuning out every scientific promise as propaganda.

The other tension here is the “celebrity scientist” effect. A famous foreign expert can bring attention, networks, maybe even better thinking. It can also become a shortcut: borrowing prestige to signal momentum. I don’t blame Ukraine for wanting allies, brains, and visibility. But the rebuild can’t be run like a conference. The winners should be the people who live with the results — engineers on the ground, local planners, utility workers, farmers, mayors — not just the people who can get a meeting announced.

If this is the start of a real pipeline from ideas to pilots to deployment, I’m for it. If it’s a one-off photo-op that lets everyone say “we’re thinking about the future,” I’m against it, because it burns time and attention — the two things post-war societies don’t have extra of.

So here’s the thing I actually want to know: will Ukraine’s reconstruction choices be driven by what looks impressive to the world, or by what can be repaired fast, governed cleanly, and trusted by the people who have to live inside it?

Frequently asked questions

What is AI agent governance?

AI agent governance is the set of policies, controls, and monitoring systems that ensure autonomous AI agents behave safely, comply with regulations, and remain auditable. It covers decision logging, policy enforcement, access controls, and incident response for AI systems that act on behalf of a business.

Does the EU AI Act apply to my company?

The EU AI Act applies to any organisation that develops, deploys, or uses AI systems in the EU, regardless of where the company is headquartered. High-risk AI systems face strict obligations starting 2 August 2026, including risk management, data governance, transparency, human oversight, and conformity assessments.

How do I test an AI agent for security vulnerabilities?

AI agent security testing evaluates agents for prompt injection, data exfiltration, policy bypass, jailbreaks, and compliance violations. Talan.tech's Talantir platform runs 500+ automated test scenarios across 11 categories and produces a certified security score with remediation guidance.

Where should I start with AI governance?

Start with a free AI Readiness Assessment to benchmark your current maturity across 10 dimensions (strategy, data, security, compliance, operations, and more). The assessment takes about 15 minutes and produces a prioritised roadmap you can act on immediately.

Ready to secure and govern your AI agents?

Start with a free AI Readiness Assessment to benchmark your maturity across 10 dimensions, or dive into the product that solves your specific problem.