Policy Advocacy - Political Research
New
We sent a six-digit code to . Enter it below — or use the link in the same email.
Or click the link in the same email — either works.
Enter the email address on your account and we'll send you a link to set a new password.
Remembered it? Sign In
Looking for an AI policy & safety job in San Francisco (SF)? Choose from 20 open roles at 7 employers. Explore real salary data and skills breakdown across remote, hybrid and on-site roles.
Free. No credit card needed.
In the week to 27 September, 18 AI policy & safety roles were open in San Francisco at 7 employers. 3 new roles were posted during the week. Of the roles that say how they work, 33% are remote or hybrid.
Among the AI policy, safety and trust roles listed on the board for San Francisco, the seniority profile sits firmly at the senior end: more than half carry a manager-level designation, close to a third are lead roles, and a small share are director positions. There are no entry-level or mid-weight roles listed. Openings are reasonably fresh, with about a third posted within the last week and another third within the past month, and none older than two months. Hiring is concentrated among a small number of employers, with the large majority of roles sitting at just a handful of companies. Evaluation is the skill mentioned most often in role descriptions, appearing in well over half of them, while mentions of LLMs, fine-tuning, and published research each appear in roughly a quarter.
Figures are measured every Monday.
AI policy and safety professionals work at the intersection of technical understanding and governance: assessing risks in AI systems, developing internal standards, engaging with regulators and external researchers, and contributing to frameworks for responsible deployment. The work often involves red-teaming, model evaluation, and producing or reviewing research that shapes how systems are used. San Francisco sits at the centre of frontier AI development, with Anthropic headquartered in the city and OpenAI maintaining a large presence in Mission Bay. Databricks and Salesforce also have their headquarters here, both building AI capabilities into their platforms. That concentration of frontier labs and enterprise AI companies means policy and safety specialists in the city are often working on problems that feed directly into live systems at scale.
Evaluation is the most commonly mentioned capability across role descriptions, suggesting candidates should be able to design and run structured assessments of model behaviour. Knowledge of large language models and experience with fine-tuning appear in roughly a quarter of descriptions. Publications and research experience feature at the same frequency, pointing to value placed on the ability to contribute to the wider field. Python, agents, and reinforcement learning appear in a smaller share of descriptions.
The employers' own advertised ranges, for individual roles at different levels — not an average and not a market rate.
On-site work is the dominant arrangement among the roles listed here, accounting for roughly two thirds of those that state a working mode. A hybrid pattern makes up most of the remainder, at just under a third. Fully remote roles are rare, representing only a small minority. For a discipline where sensitive model access and close team collaboration matter, the lean toward in-person presence is consistent with how frontier AI companies tend to organise this kind of work.
Jobs checked 4 hours ago.
Search by role, company, or anything a posting mentions.
