Context: The authors are a few EAs who currently work or have previously worked at the European Commission.
In this post, we
make the case that more people aiming for a high impact career should consider working for the EU institutions using the Importance, Tractability, Neglectedness framework, and;
- briefly outline how one might get started on this, highlighting a currently open recruitment drive (deadline 10 March) that only comes along once every ~5 years.
Why working at the EU can be extremely impactful
Importance
The EU adopts binding legislation for a continent of 450 million people and has a significant budget, making it an important player across different EA cause areas.
Animal welfare
- The EU sets welfare standards for the over 10 billion farmed animals slaughtered across the continent each year.
- The issue suffered a major setback in 2023, when the Commission, in the final steps of the process, dropped the ‘world’s most comprehensive farm animal welfare reforms to date’, following massive farmers’ protests in Brussels.
- The reform would have included ‘banning cages and crates for Europe’s roughly 300 million caged animals, ending the routine mutilation of perhaps 500 million animals per year, stopping the inhumane slaughter of a billion or so farmed fish, and reducing the overcrowding and pain-inducing growth rate of about 11 billion broiler chickens annually.’
- Still, the EU already has the highest animal welfare standards globally and is arguably the only major jurisdiction where further progress seems likely in the near term.
- The EU is also the single biggest public investor in alternative proteins, ahead of the US or any other national government.
AI safety
- The EU AI Act remains the only comprehensive AI legislation in any major global jurisdiction.
- The General-Purpose AI Code of Practice, published in July 2025, includes detailed safety and security requirements for frontier AI companies.
- Any company offering frontier AI models in the EU needs to follow the obligations in the safety and security chapter around risk assessment, external safety evaluations and incident reporting.
- The AI Office at the European Commission enforces the rules for frontier AI companies, and plays a key role in drawing them up.
Global health
- The EU and its Member States collectively account for roughly 42% of global official development assistance (ODA); amounting to approximately €96 billion in 2023.
- A significant share of this comes directly from the EU budget or is influenced by the EU. A large fraction of aid directly given by the Commission is allocated to least developed countries.
- While the UK, and especially the US, are cutting back heavily on their foreign aid and international development cooperation spending, the EU continues with significant spending in this area.
- In her 2025 State of the European Union Address, European Commission President von der Leyen unexpectedly announced a “Global Health Resilience Initiative".
- It is a signal of commitment to this area to announce such an initiative when most others are decreasing their work on global health drastically.
Biosecurity
- The Commission is a major spender on pandemic preparedness and response, having created DG HERA during COVID-19 as a dedicated institution that continues to work on the issue even as political interest in the topic quickly faded after the pandemic.
- DG HERA has a budget of ca. €1 billion per year, having spent a lot of it on stockpiling during COVID-19 and now mostly focusing on providing R&D funding for the development of vaccines, therapeutics, diagnostics and other countermeasures.
- EU funding for pandemic preparedness is becoming relatively more important as the US scales back - e.g. see the recent cancellation of nearly $500 million in BARDA mRNA vaccine contracts.
- The EU also became the first major jurisdiction to propose mandatory DNA synthesis screening in December 2025, alongside other biodefence measures in the EU Biotech Act. This might help in setting a global standard for companies screening customers for orders of synthetic DNA of dangerous viruses, raising the barrier to the development of bioweapons.
Tractability
We have and continue to observe that it is feasible to shape important policy decisions working in the Commission. While there are, of course, significant political constraints on individual policymakers and on the Commission as a whole, even relatively junior officials can make highly influential decisions.
This is particularly true for decisions that are not overtly political, but can still be extremely consequential. In practice, policymakers often have substantial discretion over technical elements of legislation and over how funding is allocated. This includes choices such as whether pandemic preparedness money is spent on measures that meaningfully reduce catastrophic risk, whether development aid goes to effective health interventions, the kinds of evaluations frontier AI companies are required to run, or which species might be prioritised for a cage ban.
It is extremely important to be in the right position at the right time, as the potential for impact within the Commission is very heavy-tailed. In some roles, officials have significant, and sometimes near-exclusive, decision power over entire chapters of legislative proposals or over tens, and in some areas even hundreds, of millions of euros in public spending each year.
On the other hand, most jobs in the Commission are fairly low impact. Surprisingly, impact or decision power does not seem that correlated with seniority, but much more with working in the exact right division of the Commission, called ‘unit’. It can be very difficult to identify units with the highest impact potential from the outside, so we would strongly recommend getting in touch with people within the Commission before deciding which unit to join. This is even more important as, in practice, there is significant path dependency from the first unit you join to future roles, even if a Commission career allows you to switch policy areas relatively flexibly in theory.
Neglectedness
EU policy remains relatively neglected among EAs - for example compared to the UK civil service, where there appears to be a community of EA-aligned policy professionals. This is particularly true for people working in EU institutions themselves, as opposed to external advocacy orgs and think-tanks. To us, working within the civil service often seems more impactful than external advocacy at the margin.
The Commission employs around 32,000 staff. Even a small increase in the number of thoughtful, impact-oriented people could shift priorities on the margin.
Our best guess is that the total number of EAs (or EA adjacent people) in the EU institutions is roughly two dozen. This varies a lot by cause area. For most of the areas mentioned above, the number of EA-motivated people is probably only around 1 or 2 (!). For global health for example, we’re currently not aware of anyone working in this field with EA motivations.
While there has been a significant cultural shift within EA towards taking policy and government roles more seriously, joining established EA organisations still feels like the default or more prestigious option. In our view, a further shift is needed, with more people feeling encouraged to join government directly and to see institutional roles as a core path to impact.
Crucially, compared to joining an EA org, your impact working in government is often entirely counterfactual. The marginal person you replace would likely have very different values and no focus on scope-sensitive impact.
Paths into the EU
UPDATE: The notice for the AD5 competition was published today - Deadline is 10 March 2026. You should fill in the initial application thoroughly, with correct dates for all work experiences and diplomas (studies, languages, etc.). You will need to submit those should you be successful.
1490 people will make it to the reserve list (from which you can be directly recruited as a lifetime civil servant), as opposed to 147 in the last such competition in 2019. Assuming a similar number of applicants as in 2019 (~25k) the probability of making it onto the reserve list would be a lot higher for this competition, around 5%. There will however likely be substantially more candidates given the exam has not taken place in a while - and it will be considerably harder to be recruited off the reserve list (around 1/3 chance, whereas in the previous system most people on the list seemed to be hired).