Job Collections

AI safety roles can be divided into four distinct categories. The deep technical roles focus on the automation and systematization of the safety checks on the AI. Management and policies focus on developing strategies for the sustainable operation of AI. Ethics examines the influence of AI from a broader societal context, such as social justice and discrimination. The operational roles help with project coordination, public relations, and administrative services.

You will need to have strong skills in mathematics, computer science, and programming, especially in Python. Understanding machine learning or deep learning principles is helpful. Moreover, Knowledge of decision theory, game theory, or causal inference is also helpful. Further, Having a PhD is an advantage, but not needed. What matters most is having the capacity to reason about complex systems and the risks they come with.

Of course you can. There are many important non-technical positions available. Such as public policy and law-trained professionals are preferred in policy and governance roles. Their ethical counterparts typically stem from philosophy or social science. Some strategists focus on public relations, aiming to articulate the importance of AI safety to the average citizen. The operations team takes charge of community outreach and other project work. The work of these staff members helps ensure that AI safety research is sociologically relevant by making it applicable in everyday life.

You can apply for apprenticeships at labs like DeepMind or Anthropic, or join programs like the AI Safety Camp hosted at different universities. Alternatively, some nonprofits offer roles in community-building and general support. A lot of self-paced online courses and open-source projects are available for skill development. There are reading groups and AI safety meetups at different locations to expand knowledge and network.