Canada and Germany Just Put $300 Million Into an AI Designed to Want Nothing
Last Updated on September 25, 2026 by Editorial Team
Author(s): Delini
Originally published on Towards AI.
Every frontier lab is racing to build systems that pursue goals. Yoshua Bengio has spent a year building the opposite, and two governments have now bet a quarter of a billion dollars that his version is the one that watches the others.
On Wednesday, 16 September 2026, at the ALL IN conference in Montréal, two government ministers stood on stage beside a 62-year-old computer scientist and announced money.
The article explains that Canada and Germany are funding Yoshua Bengio’s LawZero with up to CAD $300 million to build “Scientist AI,” a goal-free system meant to monitor other AI agents for deception and self-preservation. It argues that unlike today’s agentic models—which pursue objectives and can learn to circumvent restrictions—Scientist AI is designed to have no goals of its own, functioning more like a transparent evidence-driven monitor than a competing assistant. The piece traces how this “no-goals” approach relates to concerns about instrumental subgoals and the “papercip problem,” outlines the practical challenges (especially whether a weaker monitor can detect more capable agents), and describes what would make the idea credible: published evaluations, detection results on frontier models, real deployments by labs, and independence under political pressure. Finally, it situates the funding as an institutional bet on a third type of oversight organization (not a lab or regulator), contrasting LawZero with current methods such as self-evaluation by companies, government/standards evaluations, and adversarial red-teaming—while emphasizing that the core technical and governance gaps in AI safety are getting harder to close as capability races ahead of verification.
Read the full blog for free on Medium.
Join thousands of data leaders on the AI newsletter. Join over 80,000 subscribers and keep up to date with the latest developments in AI. From research to projects and ideas. If you are building an AI startup, an AI-related product, or a service, we invite you to consider becoming a sponsor.
Published via Towards AI
Towards AI Academy
We Build Enterprise-Grade AI. We'll Teach You to Master It Too.
15 engineers. 100,000+ students. Towards AI Academy teaches what actually survives production.
Start free — no commitment:
→ 6-Day Agentic AI Engineering Email Guide — one practical lesson per day
→ Agents Architecture Cheatsheet — 3 years of architecture decisions in 6 pages
Our courses:
→ AI Engineering Certification — 90+ lessons from project selection to deployed product. The most comprehensive practical LLM course out there.
→ Agent Engineering Course — Hands on with production agent architectures, memory, routing, and eval frameworks — built from real enterprise engagements.
→ AI for Work — Understand, evaluate, and apply AI for complex work tasks.
Note: Article content contains the views of the contributing authors and not Towards AI.
Related posts
Recent Posts
