You can always press Enter⏎ to continue

(Senior) Advisor, AI Safety

  • 1

    Pour Demain

    Pour Demain Europe

    (Senior) Advisor, AI Safety

    Brussels (Research Track: frequent trips) · Start: ASAP or to be discussed

    Frontier AI policy in Europe has risen from niche, mostly hypothetical discussions to the very top of the political agenda in recent years. The EU AI Act — the world's leading frontier model regulation — is being implemented just as serious frontier risks are beginning to emerge. Meanwhile, both European institutions and national governments are rapidly building capacity to make and act on decisions that will shape frontier AI policy for years to come. Pour Demain Europe (PDE) has built influence in this space over the past years. We are looking for a (Senior) Advisor, AI Safety to help drive our work forward, contributing to safer and more secure AI in Europe.

    → Express your interest in only 5–10 minutes with this form

    Initial deadline: 17 August 2026 · Final deadline: 31 August 2026 (23:59 CEST) · Early applications are explicitly encouraged.

    The Opportunity

    Pour Demain Europe (PDE) is scaling as AI governance shifts from drafting rooms into enforcement and implementation. We are hiring for two tracks under this role, and welcome applications for either.

    🏛️

    Policy Track (Brussels-based)

    You lead direct engagement with EU institutions, translating technical AI safety questions into concrete, politically viable policy proposals.

    🔬

    Research Track (frequent trips to Brussels)

    You develop the technical and analytical foundations behind our policy positions, from standards to evaluation methodologies to glass-box access frameworks. You do not need to be based in Brussels full time, but you engage closely and regularly with the Brussels team and with policymakers there.

    Both tracks report into PDE's AI programme and work closely with our Director of AI Policy and the wider AI team across Brussels and Switzerland.

    What Pour Demain Europe Has Already Delivered

    •Shaped key elements of the EU Code of Practice for general-purpose AI models, including serious incident reporting, third-party evaluator access beyond black-box testing, and enforcement provisions.
    •Coordinated AI safety civil society groups during key moments throughout the Code of Practice, including the drafting process itself at both technical and political levels, as well as enforcement.
    •Contributed analysis on resourcing the EU AI Office, through a dedicated report on its funding and staffing needs, alongside a joint statement with civil society, industry, and expert signatories. Our research advocacy also contributed to a European Parliament statement echoing our call, and to an AI Office hiring round opening for +40 posts.
    •Pioneered early policy work on glass-box evaluations, developing a framework for deeper, more transparent third-party access to frontier models than today's black-box testing allows.
    •Strengthened AI safety in Switzerland through direct engagement with parliament and the federal administration, and by helping shape the agenda for the Swiss AI Summit 2027.

    What You Will Do

    If you join the Policy Track

    •Represent PDE in meetings with the EU AI Office, European Commission, Parliament, and other institutions.
    •Develop and advance policy recommendations on AI Act implementation, Code of Practice enforcement, and related governance instruments.
    •Draft policy briefs, position papers, and consultation responses for policymaker and public audiences.
    •Build and maintain relationships across the Brussels AI safety and governance ecosystem.

    If you join the Research Track

    •Lead or contribute to technical research underpinning PDE's policy positions, such as evaluation access, AI standards, serious incident reporting, verification, and other risk mitigations.
    •Translate technical findings into materials usable by policymakers and civil society, in close coordination with the Policy Track.
    •Engage with external researchers, evaluators, and technical experts to keep PDE's positions grounded in the current state of the field.
    •Travel to Brussels regularly to align with the policy team and engage directly with EU stakeholders.

    In both tracks, you will work within PDE's flat, high-autonomy structure, owning your own workstreams from conception through to political or technical delivery.

    You will also contribute to PDE's broader AI strategy alongside the Director of AI Policy.

    What We Are Looking For

    Essential

    •AI Safety Experience: Professional experience in AI safety or AI governance, ideally with a demonstrated track record of concrete outputs (policy proposals, technical research, or both)
    •EU Familiarity: Familiarity with EU governance processes and institutions; direct experience with the AI Act, the Code of Practice, or comparable EU legislative processes is a strong plus
    •Communication: Ability to translate complex technical or political material for different audiences, from policymakers to technical evaluators
    •Autonomy and judgement: You organise yourself, own workstreams end to end, and take judgement calls without waiting to be tasked. You are equally willing to do the less intellectually stimulating parts of a project yourself
    •Languages: Excellent English. German or French is a plus
    •Please note: we are unable to sponsor a visa for this role. Candidates must already hold the right to work in Belgium (or the EU, as relevant)

    Desired

    •Multi-year background in AI safety governance, technical AI safety research, or EU public affairs
    •Existing network within the Brussels AI policy community or the technical AI safety community
    •Experience with OKRs, Asana, or similar systems for translating strategy into weekly execution
    •For the Research Track: hands-on experience with model evaluations, AI standards, or related technical safety work

    What We Offer

    🎯

    Direct Influence

    Shape European and global AI governance at a pivotal moment for the field

    🤝

    Substantial Autonomy

    Own your own projects, with close collaboration from an experienced, growing team

    🌱

    Flat, Fast-Moving Culture

    An evidence-driven organisation where 90% of operational decisions stay with the person doing the work

    💶

    Competitive Compensation

    Competitive salary package

    📍

    Setup

    Brussels-based for the Policy Track; flexibility in how the Research Track combines Brussels time with remote work

    Application

    Unsure if this role is right for you? If you meet many — but not all — of the essential criteria, we encourage you to apply anyway. Submit the form or reach out at info@pourdemain.ngo.

    We've kept this form intentionally brief — it takes just 5–10 minutes to submit. If you are open to either track, say so in the form; we assess candidates against both.

    Initial deadline: 17 August 2026 · Final deadline: 31 August 2026 (23:59 CEST). We review applications as they arrive, so early applications are explicitly encouraged.

    We look forward to getting to know you!

    Press
    Enter
  • 2
    Press
    Enter
  • 3
    (For Cloud-based files, please grant access authorisation for "Everyone with the link", or equivalent).
    Press
    Enter
  • 4
    Press
    Enter
  • 5
    How many years (full-time equivalent) have you worked on the following?
    1 of 5
    Press
    Enter
  • 6
    Press
    Enter
  • 7
    List 2-3 thought leaders on AI that have influenced your thinking. (Names are sufficient. A longer explanation is not necessary).
    • Huge
    • Large
    • Normal
    • Small
    Ok
    quoteCreated with Sketch.
    Ok
    Press
    Enter
  • 8
    Name your most relevant previous role in AI Safety policy. The role without explanation is sufficient. If you have worked on AI safety outside policy or policy outside of AI safety, please also list the most relevant role. Example: AI Safety Policy Advisor at X
    • Huge
    • Large
    • Normal
    • Small
    Ok
    quoteCreated with Sketch.
    Ok
    Press
    Enter
  • 9
    If you had $1 million to grant towards strengthening AI governance, which organisations would you fund, and for what work? Please give your top two ideas (excluding Pour Demain). Example: a) Organisation A for its work on X; b) Organisation B for its work on Y
    • Huge
    • Large
    • Normal
    • Small
    Ok
    quoteCreated with Sketch.
    Ok
    Press
    Enter
  • 10
    Learn more about the levels: https://en.wikipedia.org/wiki/Common_European_Framework_of_Reference_for_Languages
    1 of 4
    Press
    Enter
  • 11
    Press
    Enter
  • 12
    Press
    Enter
  • 13
    Press
    Enter
  • 14
    Ideal Setup: Brussels base, though we're open to remote arrangements for exceptional candidates willing to travel regularly to Brussels
    Press
    Enter
  • 15
    City and country. Helps us get a sense where you will be based in case of a remote arrangement.
    Press
    Enter
  • 16
    Press
    Enter
  • 17
    Press
    Enter
  • 18
    Please note: we are unable to sponsor a visa for this role. Candidates must already hold the right to work in Belgium (or the EU, as relevant).
    Press
    Enter
  • 19
    Feel free to add comments and questions. Here, you can also indicate any person who recommended you to apply (if applicable).
    • Huge
    • Large
    • Normal
    • Small
    Ok
    quoteCreated with Sketch.
    Ok
    Press
    Enter
  • Should be Empty:
Question Label
1 of 19See AllGo Back
close